Enhancing Anaerobic Degradation of Lipids in
Wastewater by Addition of Co-substrate
Yunhua Kuang
This thesis is submitted for the degree of Doctor of Philosophy
School of Environmental Science
Murdoch University
August 2002
i
DECLARATION
I declare that the work in this thesis is my own account of my research and the contents has
not been submitted for a degree at any university.
Yunhua Kuang
PUBLICATION
Y. Kuang, M. Lepesteur, P. Pullammanappallil and G.E. Ho (2002). “Influence of
co-substrates on structure of microbial aggregates in long-chain fatty acid-fed
anaerobic digesters”. Letters in Applied Microbiology. 35, 1-5
Y. Kuang, M. Lepesteur, G. E. Ho (2001). “Identification and quantification of the
anaerobic bacterial subpopulation with oligonucleotide probe hybridization under the
inhibition of long chain fatty acid” Proceeding of 9th Word Congress on Anaerobic
Digestion 2001, Antwerpen, Belgium
Y. Kuang, M. Lepesteur, P. Pullammanappallil and G.E. Ho (2002). "Enhancing the
lipids anaerobic digestion by stimulating bacteria growth with co-digestion proteins
and carbohydrates in two-phase UASB reactor" Accepted as oral presentation by VII
Latin American Workshop and Symposium on Anaerobic Digestion.
Y. Kuang, M. Lepesteur, P. Pullammanappallil and G.E. Ho (2002). “Lipid
hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems
with co-digestion with protein and carbohydrates” Submitted to Water Research
Y. Kuang, M. Lepesteur, P. Pullammanappallil and G.E. Ho (2002). “Enhanced
recovery of methanogenesis from LCFA inhibition by feeding substrates in a single
stage UASB reactor” Submitted to Bioresouce Tech.
ii
ABSTRACT
Anaerobic treatment systems are becoming increasingly popular to treat complex organic
wastes that contain carbohydrates, proteins and lipids. Lipids are widely found in sewage
and industrial wastewaters. Dairy, edible oil, fat refining, slaughterhouse, wool scouring,
meat processing plants and grease-trap wastes from restaurants generate wastewater high
in lipids. Although it is well known that lipids can be degraded by biological process,
they have been reported to inhibit anaerobic processes by causing sludge flotation and
wash-out. The inhibitory effect of lipids in anaerobic process has also been attributed to
the long-chain fatty acids (LCFAs) which are the hydrolysed products of lipids. It has
been shown that LCFA and lipids inhibit the formation of granular sludge in Upflow
Anaerobic Sludge Blanket (UASB) reactors and that the adsorption of LCFAs on to the
granules can result in its flotation and washout. It was also found that the degradation of
LCFA was very poor.
Various techniques have been employed to enhance degradation of lipids and these
include physico-chemical pre-treatment, application of two stage treatment employing
new reactor designs like Expanded Granular Sludge Bed (EGSB).
This thesis investigated the influence of co-substrates, both in the form of hydrolysed
products and polymeric form, on reducing the toxicity and enhancing the degradation of
LCFA and lipids in a single stage and two stage upflow anaerobic sludge blanket
(UASB) reactors.
The investigations were carried out on both microbiological and
physico-chemical aspects.
A combination of techniques including the use of light
iii
microscopy (LM), confocal laser scanning microscopy (CLSM), transmission electron
microscopy (SEM) and Fluorescent In Situ Hybridisation (FISH) was used to study the
characteristics of microbial aggregates and to locate microbial populations within these
aggregates. The microbial populations visualised using FISH techniques were Bacteria,
Archaea, Methanobacteriaceae, Methanomicrobiales and Methanosarcinaceae.
The
performance of digesters was also monitored by measuring bulk parameters such as
concentration of residual substrates, intermediate products (LCFAs, volatile fatty acids),
methane (or gas) production rate and chemical oxygen demand of treated effluent.
Initially batch assays were carried out to determine the effects of glucose (hydrolysis
product of carbohydrate) and cysteine (hydrolysis product of protein) on the toxicity of
sodium oleate (hydrolysis product of lipid) to methanogenesis. The results showed that
glucose and cysteine addition could reduce the toxicity of sodium oleate on the
methanogenesis and enhance the degradation of sodium oleate.
While the addition of
glucose had a better effect than cysteine on decreasing the toxicity of sodium oleate, the
combination of glucose and cysteine had the optimal result to stimulate the degradation of
sodium oleate.
Secondly the effect of addition of glucose, cysteine and sodium oleate as co-substrates on
the characteristics of granules in an LCFA fed single stage UASB were investigated. It
was shown that the addition of glucose produced the best results on the formation of
granules while both cysteine and sodium oleate adversely affected the granule formation.
In a LCFA inhibited digester glucose and cysteine addition enhanced the recoveries of
iv
different anaerobic microbial communities. Although the effects of glucose and cysteine
on the various microbial groups were different, the combination of glucose and cysteine
had the optimal results on recoveries of all bacterial groups.
The next half of the thesis investigated the influence of starch and yeast extract on the
hydrolysis and degradation of canola oil by application of one and two stage UASB
reactors. The results showed that the combined addition of protein and carbohydrate had
an optimal effect on enhancing the hydrolysis of lipid compared to the addition of only
protein or carbohydrate by promoting a balanced growth of the microbial groups. It was
also demonstrated that a two- stage UASB reactor performed better in terms of extent of
lipid hydrolysis and methanogenesis than a one-stage UASB reactor.
iv
ACKNOWLEDGEMENTS
First of all I would like to thankfully acknowledge my supervisor, Professor Goen Ho, for
his initiative to this project and for his patient supervision along the way. I would also like
to thank Dr Muriel Lepesteur-Thompson for her continuously encouragement, advice and
friendship as my co-supervisor.
I am also grateful to her most important support in
submitting this thesis.
I must express my sincere gratitude to my co-supervisor, Dr Pratap Pullammanappallil for
his valuable views and detailed comments on the thesis and five articles described in the
thesis. Without his help, the thesis would have had less to show.
A special thanks goes to Kelvin Maybury from Environmental Science for his kind technical
advice for using Gas Chromatography (GC) analysis. Douglas Clarke from Chemistry
provided technical information for analysing the lipids in the wastewater. Thanks to Gordon
Thomas from Biological Science, Peter Fallon from Veterinary Biology and Biomedical
Science for helping me to finish the preparing sludge samples and using TEM, SEM and
CLSM. I am also grateful to Dr Ralf Cord-Ruwisch for allowing me to use his chemicals
and laboratory freely.
Thanks also go to the Woodman Point wastewater Treatment Plant, Spearwood Wastewater
Treatment Plant and Maltilda Bay Brewery Company for kindly provided the different
sludge for the set up of lab-scaled experiments.
vi
I would also like to take this opportunity to say a special thank to all the staff of the School
of Environmental Science of Murdoch University. Their friendly help and expert technical
support had made my studies here very enjoyable and unforgettable.
I would like to acknowledge the Murdoch University for providing a Postgraduate
Scholarship to support me finish my PhD studies.
Last I would like to thank my husband, Dr Yun Li for his understanding and support during
my PhD studies. I am deeply indebted to my eleven-year-old daughter, Yixi and three-year–
old son, Leo, for their patience and love.
vii
LIST OF ABBREVIATIONS
ABR
anaerobic baffled reactor
CLSM
confocal laser scanning microscopy
COD
chemical oxygen demand
CSTR
continuously stirred tank reactor
EGSB
expanded granular sludge bed reactor
ELISA
immunosorbent assay
FISH
fluorescent in-situ hybridization
GC
gas chromatography
HPLC
high performance liquid chromatography
LCFA(s)
long chain fatty acid(s)
HRT
hydraulic retention time
SO
sodium oleate
SEM
Scanning electron microscopy
TEM
Transmission electron microscopy
UASB
upflow anaerobic sludge blanket
VFAs
volatile fatty acids
viii
TABLE OF CONTENTS
Abstract …………………………………………………………………………………...iii
Acknowledgments ……………………………………………………….………………..vi
List of abbreviations……………………………………………………………………..viii
Chapter 1 Introduction…………………………………………………………………..1
1.1
General introduction………………………………………………………….1
Chapter 2 Literature review…………………………………………………………….5
2.1
Lipid anaerobic digestion process…………………………………………...5
2.2
Microbiology of anaerobic digestion……………………………………….. 9
2.3
The inhibition of LCFAs on anaerobic microoganism…………………….27
2.4
Anaerobic treatment of lipids containing wastewater……………………..39
2.5
Difficulties and new approach of lipids anaerobic treatment……………..40
2.6
Research objective………………………………………………………… 46
2.7
Outline and outcomes of the thesis………………………………………… 47
Chapter 3
Materials and methods………………………………………………49
3.1
Experimental lab-scale reactors and set up………………………………...49
3.2
Batch experiment for LCFA toxicity test…………………………………..55
3.3
Quantification and identification of anaerobic bacteria groups by
oligonucleotide probe hybridization…………………………………………56
3.4
Total cell counts with DAPI method…………………………………………59
3.5
Investigation of granular structure with scanning and transmission electron
microscopy……………………………………………………………………. 60
3.6
Confocal laser scanning microscopy (CLSM) ………………………………61
3.7
Analysis of gas and effluent of batch and UASB…………………………….63
3.8
Measurements of lipids, carbohydrates and proteins and methods for
measurements of COD, total suspended solids (TTS), volatile solids (VS),
suspended solids (SS) and volatile suspended solids (VSS). ……………….65
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of cosubstrates in batch tests…………………………………………………………..68
4.1
Introduction…………………………………………………………………..68
4.2
Material and Methods……………………………………………………….70
4.3
Results and Discussion……………………………………………………… 73
4.4
Conclusion…………………………………………………………………….84
Chapter 5 Influence of co-substrate supplementation on granule characteristics
in a LCFA fed single stage anaerobic digester……………………………..85
5.1
Introduction…………………………………………………………………..85
5.2
Materials and Methods………………………………………………………87
5.3
Results and Discussion……………………………………………………….90
5.4
Conclusion………………………………………………………………101
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding
glucose and cysteine in a single stage UASB reactor…………………102
6.1
Introduction……………………………………………………………...102
6.2
Materials and Methods………………………………………………….104
6.3
Results and Discussion…………………………………………………..105
6.4
Conclusion………………………………………………………………..118
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate
and protein in a one-phase UASB reactor by monitoring the microbial and
physico-chemical indicators……………………………………………119
7.1
Introduction……………………………………………………………..119
7.2
Materials and Methods…………………………………………………122
7.3
Results and Discussion………………………………………………….125
7.4
Conclusion………………………………………………………………..137
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic
treatment systems……………………………………………………….138
8.1
Introduction………………………………………………………………138
8.2
Materials and Methods…………………………………………………..140
8.3
Results and Discussion…………………………………………………...143
8.4
Conclusion…………………………………………………………………162
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth
with co-digestion of proteins and carbohydrates in two-phase UASB reactor
9.1
Introduction………………………………………………………………….163
9.2
Materials and Methods……………………………………………………...165
9.3
Results and Discussion………………………………………………………167
9.4
Conclusion…………………………………………………………………….179
Chapter 10. Conclusion and recommendation
10.1
Conclusion……………………………………………………………………180
10.2
Recommendation for future works…………………………………………183
REFERENCE……………………………………………………………………………185
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 1: Introduction
Chapter 1 Introduction
1.1 General introduction
Lipids, which can be traditionally characterised as fats, greases and oils, are widely
found in industrial and municipal wastewaters. Lipids in industrial wastewaters are
mainly esters of straight-chain, even-numbered long chain fatty acids and linear polyols
(triglycerides or phospholipids) and their hydrolysed products (Rinzema, 1988).
Physicochemical techniques, such as gravity separation, flotation and emulsion, can
effectively treat wastewaters containing lipids. However, even after these processing
steps, considerable amount of emulsified or colloidal lipids is left in the wastewater
(Rinzema, 1988). Thus, to completely remove the lipid remaining, further treatment is
required. Biological techniques, with their low requirement for expensive chemicals or
energy and with fewer harmful end products, have been increasingly used in recent
years.
Biological treatment of lipids may be accomplished by aerobic or anaerobic digestion.
Anaerobic digestion is the biological degradation of organics into carbon dioxide and
methane, while aerobic digestion oxidises organics to carbon dioxide and water using
oxygen. Anaerobic treatment has advantages over aerobic treatment in that there is less
energy required and the end product (methane) can be used for energy production.
Reactors can be easily restarted even after several months of shut down (Beccari et al.,
1996). However, there are some disadvantages to this method when compared to
aerobic digestion. Anaerobic treatment is more susceptible to failure from changes in
reactor conditions, as well as from difficulties during reactor operation even under
stable conditions (Fongastitkul et al., 1994). Other drawbacks to the use of anaerobic
1
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 1: Introduction
digestion include a longer retention time and the comparative lack of development of
lipid treatment processes (Rinzema, 1988). Furthermore, there is a lack of
understanding of the identity and complexity of the microorganisms involved in this
process due to the limitation of traditional culture-based methods.
With the traditional culture based methods, only a relatively small fraction of the
microorganisms making up a natural community can generally be cultured, and these
cultured microoraganisms are also difficult to be identified. In recent years, groupspecific 16S rRNA hybridization probes have been used to study the microbial
populations in anaerobic digesters (Amann et al., 1990a; Raskin et al, 1994a; Stahl, et
al. 1988). This technique is attractive due to the specificity of the probes for both
cultured and uncultured microbial cells in natural complex microbial communities
(Raskin et al., 1995).
In anaerobic digestion, following the hydrolyis of LCFAs from lipids, LCFAs are
further degraded, mainly by β-oxidation, to acetate and hydrogen and finally converted
to methane or carbon dioxide by methanogenesis (Komastsu et al., 1991). However,
even low concentrations of LCFAs can be inhibitory to the process (Hanaki, 1981;
Koster, 1986; Angelidaki et al., 1990). It was suggested that LCFA are toxic to
bacteria because of their damaging effect on the transport channel of bacterial cells,
inducing the lysis of protoplasts (Galbraith, et al., 1971; Galbraith and Miller, 1973).
Considerable work has hence been carried out on decreasing the toxicity of LCFAs on
anaerobic microoganisms (Koster, 1986, 1987; Hwu, 1996; Rinzema et al., 1993).
These studies were limited to the addition of chemicals and the improvement of reactor
technologies and the performance was monitored by indirect parameters, such as
2
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 1: Introduction
methane production rate, chemical oxygen demand (COD), biological oxygen demand
(BOD) and volatile fatty acids (VFA). The effects of LCFAs directly on microbial
groups have not been extensively investigated.
Furthermore, although the toxicity of LCFA on anaerobes is well defined, effective
methods that are capable of decreasing the inhibition of LCFA are poorly documented.
During anaerobic lipid digestion, many factors can affect the growth of microbial
communities and interfere with this process.
An important factor is the nutrient
composition. Ahring et al. (1992) observed that supplementing co-digested lipid waste
with other industrial waste enhanced the digestion process. Angelidaki and co-workers
(1998) also investigated the effect of co-digestion of complex substrates by using
mathematical modelling. Mousa (1998) reported that the presence of trace organics
reduced the inhibition of LCFA. The complex microbial communities in this process
were not identified. In depth research is clearly necessary to identify and understand
the effect of various substrates on the shifts of microbial subpopulations which related
to the activities of anaerobic digestion of lipids.
Research has also been conducted on the selection of the optimal pH, temperature,
substrate components and loading rate for the anaerobic reactor without
characterisation of the real structure of microbial community inside the reactor (Koster,
1987; O'Flaherty et al., 1997). The effect of altering a single variable of digester
operation is difficult to ascertain by monitoring bulk parameters. Any effect observed
will not be a direct result but one arising from the effect the change has, in conjunction
with other factors, on the population dynamics of the digestor microbial community.
The complex microbial subpopulations which are involved in the degradation of lipids
3
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 1: Introduction
in anaerobic digestion are the key factors to maintaining the anaerobic mineralisation
process. Knowledge of the presence of different subpopulations during LCFA and lipid
anaerobic digestion is very limited due to the fact that traditional techniques are unable
to completely identify and quantify the heterogeneous microbial community in
anaerobic digestion. Methanogens represent one of the most difficult groups amongst
this community to investigate with culture-based methods even though they constitute
an important group in forming the final product.
4
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Chapter 2 Literature review
2.1 Lipid anaerobic digestion process
Anaerobic digestion consists of a series of microbiological processes that convert organic
compounds to methane in the absence of oxygen. A schema of these processes is displayed
in figure 2.1:
Lipids
H Y
LCFAs
Proteins
D
R O L
Glycerol
A
β-OXIDATION
Carbohydrates
Y
S
I
S
Amino acids
C I D O G E
Sugar
N
E S I S
Volatile Fatty Acids (C>2) D
ACETOGENIC OXIDATION
Hydrogen
Acetate
HYDROGENOTROPHIC
METHANOGENESIS
ACETOTROPHIC
METHANGENESIS
Methane
Fig. 2.1 Pathway for the metabolism of organic compound in anaerobic digestion
The overall reaction for this process can be described as:
Organic materials + H2O
biomass + CH4+ CO2 + NH4 +
5
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Anaerobic processes occur naturally in the environment, such as in ocean sediments, lakes,
river, swamps, soils and the gastrointestinal tract of animals. As one of the oldest forms of
biological wastewater treatment, anaerobic digestion has been used for more than a century by
human civilisation. The great quantities of organic waste generated by urban and industrial
activities and have resulted in a reawakening of interest in the anaerobic process as a means of
disposal.
During anaerobic digestion, organic material is converted to biomass, methane and carbon
dioxide. Nitrogen is released as ammonia and only very little of it is used for anaerobic
bacterial growth. In this process, different kinds of microbial populations are involved in the
degradation of organic compounds into methane and carbon dioxide, which include several
consecutive and parallel reactions (Fig. 2.1).
In the first step, complex organic compounds (proteins, carbohydrates and lipids) are
hydrolysed into simpler compounds by extracellular enzymes that are excreted by
fermentative bacteria. Proteins are degraded via (poly)peptides then to amino acids using
proteases; carbohydrates are converted to soluble sugars (saccharides) using cellulases; and
lipids are transformed into long chain fatty acids (LCFAs) and glycerol using lipases.
Pavlostathis and Giraldo-Gomez (1991) suggested that lipid hydrolysis is generally faster than
protein or carbohydrate hydrolysis, and considered lipid hydrolysis to be a rapid process in
anaerobic digestion (Hanaki, 1981; Angelidaki et al., 1995).
Some researchers reported
however, that lipid hydrolysis was the rate limiting step in digestion (Beccari et al., 1996).
Other research indicated that lipid hydrolysis was not usually the rate-limiting step and the
overall conversion rate was limited either by the degradation of LCFAs or by the physical
6
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
processes of dissolution and mass transfer of these acids (Rinzema et al., 1993; Hanaki et al.,
1987). The apparent discrepancy between the results obtained by different researchers may be
attributed to differences in feed concentration and physico-chemical conditions in their
experiments.
As the first step of anaerobic digestion, hydrolysis is a very important factor. Although some
work has been carried out in this area, the focus was primarily on the changes of influents and
effluents with mathematical modelling. The most commonly applied model for lipid as well
as carbohydrates and protein hydrolysis is first-order kinetics (Pavlostathis and GiraldoGomez, 1991). However, Two-phase and Contois were approximately equivalent and best
simulated the hydrolysis data (Batstone, 1999).
Rinzema (1993) suggested that higher
biomass concentration was necessary for hydrolysis.
Little is known however, about the microbial communities involved in lipid anaerobic
hydrolysis. After hydrolysis, LCFAs and glycerol undergo fermentation or acidogenesis.
Because LCFA require an external electron acceptor for oxidation, degradation of LCFA
happened during acetogenesis. While mainly saturation or hydrogeneation of unsaturated
LCFAs occurs in this process, glycerol is also mainly degraded to acetate, lactate and 1,3propanediol (Batstone, 2000).
Syntrophic acetogenesis commences follows hydrolysis and fermentation. During this period,
the products of acidogenesis are converted into substrates i.e. acetate, hydrogen and carbon
7
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
dioxide for methanogenesis. The reaction can be represented by the following equation
(Haandel et al., 1994):
CXHyOz + 1/4 (4x-y-2z)H2O →1/8(4x+y-2z)CH3 COOH+ 1/4(2z-y)CO2 (when y<2z )
Or
CXHyOz +(x-y) H2O → x/2 CH3 COOH + 1/2(y-2z)H2 (when y>2z).
Normally all LCFA and ethanol require syntrophic acetogenesis before methanogenesis along
two pathways. β-oxidation is the main pathway observed in LCFAs anaerobic digestion:
LCFAs
activation
LCFA-CoA
saturation
saturation LCFAs
β-oxidation
7 repetitions of β-oxidation
8 acetic acid
Palmityl-CoA + acetyl-CoA
acetic acid
ω-oxidation is the another pathway to LCFA oxidation by carboxylation of the far methyl
carbon.
The final step is methanogenesis which is usually the rate limiting step in the overall anaerobic
process of soluble substrates. Methanogenesis consists of acetotrophic and hydrogenotrophic
methanogenesis.
In acetotrophic methanogenesis, methane is produced from acetate by
acetotrophic bacteria, while in hydrogenotrophic methanogenesis, the hydrogenotrophic
bacteria covert carbon dioxide and hydrogen into methane.
8
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
2.2
Microbiology of anaerobic digestion
The biological systems involved in anaerobic digestion have been studied for more than 80
years. Even since the first description of the degradation of cellulose by methane formation
in 1906, microbiological studies of anaerobic bacteria have been proved to be difficult due
to the complexity of the anaerobic metabolism (Hwu, 1997). Bryant (1972) and several
other researchers used traditional laboratory techniques to isolate and study anaerobic
bacteria. These researches showed that anaerobic ecosystems involved complex interactions
between microoganisms. It was also found that different steps in the anaerobic process were
dominated by different bacterial groups.
In the last 20 years, several biochemical and molecular methods have been developed to
characterise the microbial communities involved of anaerobic digestion particularly in
anaerobic reactors such as continuously stirred tank reactor (CSTR), anaerobic baffled
reactor (ABR), upflow anaerobic sludge blanket reactor (UASB) and expanded granular
sludge bed reactor (EGSB).
2.2.1 Molecular techniques
The molecular methods, that have been applied to anaerobic digestion were primarily
immunotechniques and hybridisations with oligonucleotide probes targeting the small subunit
ribosomal RNA.
9
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Immunological tests for characterization of anaerobic bacteria were first introduced by
Conway and Macario (1982) and Archer (1984).
These techniques involved the use of
polyclonal antibodies which were initially developed for use in defined cultures (Conway and
Macario, 1982). This technique has not been developed further due to limitations such as the
requirement for pure strains and time consuming production of antibodies. Nevertheless, the
utilisation of the enzyme linked immunosorbent assay (ELISA) to methanogens (Archer, 1984
; Sorensen et al., 1997) has proved to be highly reliable.
The use of oligonucleotide probes for characterization of anaerobic bacteria was first
introduced by Amann and co-workers (Amann et al., 1990a; Amann et al., 1990b) and was
further developed by Raskin and co-workers (Raskin et al., 1994a; Raskin et al., 1994b;
Raskin et al., 1995 ). The method was demonstrated to be a highly specific and reliable tool in
detection and identification studies. It has been widely applied in the characterization of
methanogens, acetogens and sulfate reducing bacteria (SRB) (Amann et al., 1992; Kane et al.,
1993; Santegoeds et al., 1999).
Phylogenetically defined groups of bacteria can be quantified and visualized by hybridisation
of the
32
P –and by hybridisation of fluorescent-dye-labelled oligonucleotide DNA probes to
the 16S rRNA from samples respectively. With this technique, the anaerobes could be studied
in their natural environment. The limitation of traditional culture-based techniques, in which
only a small fraction of microorganism can be cultured due to the limitation of condition is
thus avoided. These microbes that could be cultured were often difficult to quantify and
identify. However, the use of fluorescent dye or enzyme-conjugated probes for hybridisation
10
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
with specific RNA in whole cells provided clear visualisation of single cells in the natural
population. When this technique was used in conjunction with PCR, cloning, and comparative
sequencing of the cloned DNA, identification and quantification of uncultured or undescribed
strains was also possible. An alternative quantification method for using a 32P –labelled probe
hybridization was shown to involve extensive work, such as RNA extractions, slot-blot
hybridisations and data analyses. Although the results of this method were reliable, precision
and detectability should be checked when examining species present in low numbers in the
natural environment (Raskin et al., 1996).
The
oligonucleotide
approach
is
more
attractive
than
immuno-methods
because
oligonucleotide probes can be designed from DNA or RNA sequences obtained directly from
environmental samples and then synthesised on the DNA synthesizer.
The production of
antibodies required for immuno-methods involve more labour and the use of animal hosts.
Furthermore, immuno-methods require the isolation of an immunising strain from a defined
culture. The antibodies raised in animals could be to a specific strain, and might not be similar
to the majority of this kind of organisms in the environment, and thus they could not detect
most of the organisms. These limitations reduce the general applicability immuno-methods in
microbial environmental studies.
2.2.2 Taxonomy of anaerobic bacteria
Classification of anaerobic bacteria can be described by two systems (Fig. 2.2). The first
system relies on morphological and physico-chemical characteristics, such as the fivekingdom phylogeny system from a classical taxonomy, in which anaerobes are in both
11
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
kingdoms of the Prokaryotes - the Archaea and the Bacteria (Woese et al., 1990). Fungi,
Animals and Plants are another three kingdoms that belong to Eucaryotes. The Archaea
include the methanogens and a few thermophilic sulphur-metabolises (Mah et al., 1986)
while the other obligate anaerobes are Bacteria (Table 2.1). The other classification is based
upon molecular structure and 16S rRNA sequences (Woese et al., 1990) and all living things
were included in three domains. Using this method, it was found that anaerobes were
present in the Bacteria and Archaea domains. This latter form of classification has been
widely used in recent anaerobic research (Zheng et al., 1996; Sekiguchi et al., 1999). It
appears to provide a more specific system in the highest levels of natural classification.
Although some fungi and protozoa may be involved in anaerobic reactions, anaerobic
bacteria are the dominant microorganisms in anaerobic digesters.
Fermentative bacteria, acidogenic bacteria, acetogenic bacteria and methanogenic archaea
are the four main bacteria groups that are involved in reactions and operate synergistically.
Sulfate-reducing bacteria however, can also act either as acetogenic bacteria or as bacteria
that compete for nutrients with methanogenic bacteria (Raskin et al., 1996).
Fermentative bacteria are responsible for the two first stages of anaerobic digestion
(hydrolysis and acidogenesis). Fermentative bacteria, also called hydrolytic bacteria, break
down complex organic molecules into soluble monomer molecules. The function of
fermentative bacteria in the anaerobic process depends on both the initial substrate and the
environmental conditions (Novaes, 1986).
12
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Fig. 2.2. Tree of life and the situation of three domain (Adopted from Norman, 1997).
13
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
ARCHAEA (arkeootit; archbacterians, archeans)
|-- KORARCHAEOTA
|--o CRENARCHAEOTA
| |--+-- Thermofilum
| | `--+-- Pyrobaculum
| | `-- Thermoproteus
| `--+--+-- Sulfolobus
| | `--+-- Metallosphaera
| | `-- Acidianus
| `--+--+-- Pyrolobus
|
| `--+-- Hyperthermus
|
| `-- Pyrodictium
|
`--+--+-- Thermodiscus
|
| `-- Igneococcus
|
`--+-- Staphylothermus
|
`--+-- Desulfurococcus
|
`-- Thermosphaera
`--o EURYARCHAEOTA
|-- Thermococcales
|-- Thermoplasmales
|-- Archeoglobales
|-- Methanopyrales
|-- Methanococcales
|-- Methanobacteriales
`--+-- Methanomicrobiales
`-- Halobacteriales
Table 2. 1 The classification of Archaea (2003)
(http://www.palaeos.com/Kingdoms/Prokaryotes/Archaea.htm)
14
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
2.2.2.1 Fermentative bacteria
When fermentative bacteria are involved in hydrolysis, different extracellular enzymes such
as cellulase, proteases and lipases are secreted (Batstone, 2000). These are released from the
cell membrane and break down the organic materials outside the cell. Enzyme secreting
rates is usually assumed to be related to microbial growth rate and as well as environmental
factors. The mechanism of the reaction between enzymes and substrates is still not very well
understood.
Although fermentative bacteria are very important in anaerobic digestion, the conditions
required for their growth, as well as cell metabolism and ecology are not completely
understood. Toerien and Hattingh (1969) reported that species belonging to the genera of
Bacteroides, Clostridium, Butyrivibrio, Eubacterium, Bifidobacterium and Lactobacillus are
dominantly found in the hydrolysis process.
Gram-negative
bacteria,
while
Of these, Bacteroides and Butyrivibrio are
Clostridium,
Eubacterium,
Bifidobacterium
and
Lactobacillus are Gram-positive.
Acidogenesis takes place intracellularly where the soluble protein, sugars and LCFA are
degraded to short-chain organic acids and alcohols. Lipases convert lipids to long-chain
fatty acids. A population density of 104 - 105 lipolytic bacteria per ml of digester fluid has
been reported (Zeikus,1980). Clostridia and the micrococci appear to be responsible for most
of the extracellular lipase production.
Although the technological aspects of acidogenesis
15
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
have received much attention, little is known about the microbial diversity and community
structure required for it to occur (Britz et al., 1994).
Britz et al. (1994) isolated 288 acidogenesis strains from the liquid phase of four anaerobic
digesters, which were fed with different substrates and kept in different physico-chemical
conditions. Clostridium bifermentans was the predominant species in four digesters, while
Bacillus, Citrobacter,
Enterobater,
Fusobacterium,
Pseudomonas,
Klebsiella
and
Staphylococcus were the most numerous species in the other digesters.
So far, the research on fermentative bacteria has been conducted using traditional
microbiological approaches. Although both are known as fermentative bacteria, distinctions
and subsequent classifications between the hydrolytic and acidogenic bacterial groups are
still not clear. The mechanisms used by these bacteria are even less understood. There
would hence appear to be a need for the application of molecular techniques to study these
bacteria in greater detail. Van Dyke and McCarthy (2002) had applied molecular techniques
to check Clostridium populations.
2..2.2.2 Acetogenic bacteria
Acetogenic bacteria can be divided into hydrogen producing acetogenic bacteria and
homoacetogenic bacteria. Their main function in anaerobic digestion is to convert fatty
acids and alcohols into acetate, hydrogen, and carbon dioxide. These are subsequently used
by methanogens for their metabolism.
16
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Hydrogen producing acetogenic bacteria are essential for the anaerobic digestion because of
their action in catabolizing propionate and other organic acids into acetate and hydrogen. A
few species have been isolated and some information is known about their main nutritional
requirements (Bryant et al., 1967; Boone et al., 1980; McInerney et al., 1981).
The homoacetogenic bacteria are also important in anaerobic digestion and are characterized
as chemolitotrophic, H2 and CO2 users. It was reported by Bryant et al. (1967) that
methanogenic H2 consuming bacteria and an acetogenic bacteria called “organism S” had a
syntrophic association. The methanogenic H2 consuming bacteria used the H2 released from
the organism, with the maintenance of low levels of H2 as a result. Boone and Bryant (1980)
reported a similar action between acetogenic bacteria Syntrophobacter wolinii and H2 –
using bacterium Desulfovibrio in which propionate was degraded into acetate, CO2 and H2.
McInerney et al. (1981) showed that the rumen bacteria Syntrophomonas wolfei and rumen
methanogens formed a structural and metabolic association enabling them to β-oxidize fatty
acids to acetate and H2 or acetate, propionate and H2.
Although acetogenic bacteria play an important role in LCFA degradation, only four
syntrophic acetogens have been described so far (Roy et al., 1986; Lorowitz et al., 1989;
Angelidaki et al., 1995; Svetlitshnyi et al., 1996). Of these four syntrophic acetogens,
Syntrophomonas sapovorans
and Syntrophomonas wolfei sub.sp. saponavida
were
mesophilic, while Thermosyntropha lipolytica and a short rod thermophile were
thermophilic.
17
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
With traditional culture methods, the syntrophic partners of Syntrophomonas sapovorans
and Syntrophomonas wolfei sub.sp. saponavida have been isolated. These microbes were
identified
as
Methanospirillum
hungatei
and
Desulfovibrio
sp.
respectively.
Methanobacterium sp was identified as the syntrophic partner of Thermosyntropha
lipolytica while Methanobacterium thermoautotrophicum was the syntrophic partner of the
short rod thermophile.
Current developments in biological science using molecular approaches have widened the
scope for studies on the quantification and identification of the natural ecology of acetogenic
bacteria (Hansen et al., 1999; Zhao et al., 1993).
Hansen et al. (1999) used new
oligonucleotide probes to characterize the phylogenetic groups of mesophilic members of
the family Sytrophomonadaceae as well as their syntrophic relationship with the
hydrogenotrophic methanogens.
2.2.2.3 Archaea
Compared to the other three groups described, methanogenic bacteria have been well
studied. The methanogenic group is composed of both gram-positive and gram-negative
bacteria with a wide variety of shapes. A number of specific characteristics are present that
make this group quite different from other bacteria and cause their inclusion in the domain,
Archae (Woese et al., 1990). Methanogens do not have murein in their cell walls and are
related to the most primitive microorganism groups while it is common in Bacteria.
18
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
In
1936,
Barker
initially
classified
methanogens
into
only
one
family,
the
Methanobacteriaceae, which consisted of four genera: Methanobacterium, Methanosarcina,
Methanococcus, and Methanospirillum. Based on the 16S rRNA structure, Balch et al.
(1979) proposed a new taxonomic classification (table 2.2).
ORDER
Methanobacteriales
FAMILY
Methanobacteriaceae
GENUS
Methanobacterium
Methanobrevibacter
Methanosphaer
Methanothermaceae
Methanothermus
Methanococcales
Methanococcaceae
Methanothermus
Methanomicrobiales
Methanomicrobiaceae
Methanomicrobium
Methanogenium
Methanoculleus
Methanospirilum
Methanocorpusculaceae
Methanocorpusculum
Methanosarcinaceae
Methanoplanus
Methanoplanaceae
Methanosarcina
Methanococcoides
Methanolobus
Methanohalophilus
Methanosaeta
Table 2.2 Classification of methanogens
19
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Methanogens grow slowly in wastewater and their generation times range from 3 days at
35ºC to as high as 50 days at 10ºC. There are two main subcategories, hydrogenotrophic
methanogens and acetotrophic methanogens.
Hydrogenotrophic methanogens convert
hydrogen and carbon dioxide into methane.
CO2 + 4H4→ CH2 + 2H2O
Although it was originally considered that hydrogenotrophic methanogens only used
hydrogen as the electron carrier, evidence has been presented that in some cases formate can
act as an electron carrier (Boone et al., 1989).
Acetotrophic methanogens convert acetate into CH4 and CO2. It is this step that is thought
to be the bottle-neck of the whole anaerobic digestion process.
This is because the
acetotrophic methanogens are slow-growing and do not adapt well to environmental
changes.
CH3COOH → CH4 + CO2
Methanosarcina and Methanothrix/Methanosaeta are the only two genera that have been
confirmed so far to use acetate.
About two thirds of the methane produced is derived from the conversion by acetotrophic
methanogens. The other third is the result of carbon dioxide reduction by hydrogenotrophic
methanogens.
20
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Species differ in the range of environmental conditions within which they can compete
advantageously. Methanogens can develop in an environment where the redox potential is
approximately -300 mV with a temperature range of 15-40ºC.
Methaobacterium thermoautotrophicum
can grow above 60ºC.
A few species such as
The optimal pH for
methanogens is between 6.5-8.0.
While the physico-chemical and biochemical mechanisms of methanogens have been widely
studied, the ecology of microbial populations and communities in natural anaerobic
digesters remains largely unexplored (Raskin et al., 1994a). The recent development of
molecular techniques will allow appropriate new methods to be applied to the research of
methanogens.
2.2.2.4
Sulfate reducing bacteria
Sulfate reducing bacteria (SRB) are widely found in anaerobic digesters coexisting with
methanogens and acetogens. SRB produce acetate, H2 and sulfide (which are used by
methanogens) and will sometimes compete for intermediates, such as hydrogen, formate,
acetate, propionate and butyrate with methanogens or acetogens (Isa et al., 1986). The
activity of SRB is very complicated. The syntrophic and competitive interactions of SRB
with methanogens and acetogens have been widely studied although some aspects remain
unclear.
21
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
The genera of SRB have been defined on the basis of morphology rather than on
physiological grounds (Table 2.3). They include Desulfovibrio, Desulfotomaculum and
Desulfobulbus which are those principally found in anaerobic digesters (Amann et al., 1992;
Kane et al., 1993; Santegoeds et al., 1999). The results of molecular research of these
genera have provided some valuable additional information on SRB. Kane et al. (1993)
used oligonucleotide probes and polymerase chain reaction (PCR) to isolate and identify the
SRB species successfully. Using similar techniques, Amann et al. (1992) suggested that the
SRB were involved in developing and establishing biofilms. Using molecular methods and
microsensors, Santegoeds et al. (1999) studied the relation of SRB and methanogens in
anaerobic aggregates. Devereux and his colleagues (1992) designed genus- and groupspecific hybridisation probes for SRB. These probes were also used by Raskin (1996) and
co-workers in their study of the competition and coexistence of SRB and methanogenic
populations in anaerobic biofilms. The results of their work showed that 15% of SRB were
present in the methanogenic reactor even in presence of sulfate and SRB might functioned as
proton-reducing acetogens and/or fermenters.
The increase of sulfate in the anaerobic
digester could significantly increase the SRB population level and decreased the
methanogen population level in the reactor, indicating that the addition of sulfate benefited
the competition of SRB with methanogens. However, they also found that some SRB
(Desulfococcus/Desulfosarcina/Desulfobotulus groups) were more competitive without
sulfate.
22
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Genus
Morphology
Motility
flagella
Spores
Desulfovibrio
vibrios
+
single, polar
-
Desulfotomaculum
rods
+
peritrichous
+
Desulfobacter
rods/cocco-bacilli
-
-
-
Desulfococcus
cocci
-
-
-
Desulfosarcina
irregula, in packets
-
-
-
Desulfobulbus
citron-shaped cocci
-
-
-
Desulfonema
filaments
-
-
-
Table 2.3 Traditional classification of SRB (Levett, 1990).
2.2.3 Anaerobic bacteria and granulation
Microbial aggregation in anaerobic reactors is essential in biological wastewater treatment.
Aggregation significantly enhances the settle ability of biomass and subsequent bacterial
retention in the reactor. Granulation occurs due to microbial immobilisation which leads to
aggregation and growth in anaerobic reactors. It has been reported that syntrophic bacteria
and Methanosaeta spp. play an important role in the anaerobic granulation process (Elmamouni et al., 1997).
The species Methanosaeta spp. (former Methothix) and
Methanosarcina spp. have been identified as the main methanogens in granulation (Schmidt
et al., 1996).
23
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Granules are usually spherical with a diameter ranging from 0.14 to 5 mm (Schmidt et al.,
1996).
Granules are typically developed from inorganic components and extracellular
polymers (ECP) produced by bacteria. The inorganic minerals content in granules was
found to range from 10 to 90% of the granular dry weight while the ECP was between 0.6 to
20% in volatile suspended solids (Schmidt et al., 1996). From microscopic observations, it
would appear that bacterial ECP were more important than inorganic components in the
structure and maintenance of granules. Investigation of the granulation also showed that
methanogens were more important than acidogens and acetogens in the granulation process.
Many theories regarding anaerobic granulation have been proposed (Macleod et al., 1990;
Fang et al., 1994a). Macleod and Costerton (1990) suggested that the structure of bacterial
aggregates in UASB granules was composed of three layers: an exterior layer of acidogens
and H2 consuming organisms, a middle layer dominated by H2 producing acetogens and H2
consuming organisms, and a third layer consisting of Methanosaeta - like species and large
cavities. There were conflicting observations however, on the different substrates forming
the granular bacterial structure. Fang et al. (1994b) suggested that the typical three layered
granule was limited to full-scale anaerobic reactors fed with soluble carbohydrates. No
layers were evident in the granules from a reactor degrading non carbohydrate substrates.
High rate organic waste treatment anaerobic reactors are more likely to fail due to the
presence of lipids and LCFA inhibiting the growth of acidogens, acetogens and
methanogens as well as the wash-out of the sludge. Although lipids were shown to enhance
biogas production significantly (Angelidaki et al., 1990), the high potency of this substrate
24
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
could easily lead to an overloading of the reactor. Furthermore, a long adaptation period for
microbial communities in the reactor would be required. Achieving granulation in anaerobic
reactors treating lipids and LCFAs has proven difficult. Some work has been undertaken in
this field but focussed mainly on the engineering and physico-chemical aspect of the
process. Little study has been conducted on the effect of substrates on granulation. Samsoon et al. (1991) observed that granulation did not take place when oleate, a typical LCFA
monomer of lipids, was used as the sole substrate for a UASB system. The result from
Batstone (2000) demonstrated that the microbial “layers” were not correct.
2.2.4 Physicochemical factors affecting anaerobic bacteria
Considerable research has been conducted on the many environmental physico-chemical
factors related to the growth of anaerobic microorganisms. Physical factors such as
temperature, sludge retention time (SRT), hydraulic retention time (HRT), loading rate and
reactor design has been widely investigated. Anaerobic digestion has been shown to operate
under two temperature ranges: mesophilic conditions (25-40ºC) and thermophilic conditions
(> 45ºC). Zinder et al. (1984) reported that the production of methane increased with the
temperature increase.
When the temperature increased to 70ºC, methane production
decreased. Other methanogens and bacterial groups were shown to differ in optimum
growth temperatures (Angelidaki et al., 1995; Hwu, 1997). Differences in pH for optimum
growth were also apparent. Methanogens required pH values in the neutral range, 6.6-7.6
while the optimum pH for acidogens was 5.5-5.9 (McCarty and Mosey, 1991; Toerien and
Hattingh, 1969). Because of the different temperature and pH requirements of acidogens and
25
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
methanogens, a two- phase reactor was designed for separating acidogenesis and
methanogenesis.
Chemical factors, organic nutrients such as lipids, hydrocarbon and protein, and other trace
metals are very important in anaerobic digestion. The knowledge of the nutrient demands of
anaerobic bacteria is poor. However, most research on nutrients in anaerobic digestion
focused on nutrient removal during the process. The nutrient requirement for the growth of
anaerobic bacteria has not been extensively examined.
Nutrients are a key factor in
anaerobic degradation due to their important function for bacterial growth. Although it can
be used as an important energy source if operated successfully, high concentrations of
organic matter can result in the reactor inhibition (Becarri et al., 1996). Due to their high
energy content, lipids and associated products of hydrolysis, LCFA are also an attractive
potential substrate. However, lipids are very difficult to hydrolyse and can result in the
flotation and washout of granular sludge and biomass, while LCFAs inhibit anaerobic
microorganisms even at low concentrations (Rinzema, 1988; Hwu, 1997).
Most of the research on the physico-chemical factors affecting on anaerobic digestion has
been observed using traditional indicators such as volatile fatty acids (VFA), methane
production, specific methanogenic activity (SMA), F420 and ATP (Tenorrio, 1995). The
effects of microbial population structure on the anaerobic digestion are not fully known.
The lack of microbial studies on the anaerobic digestion of lipid is due to the limitation of
traditional techniques, such as selective enrichment and pure-culture isolation. Natural
microbial ecosystems are very difficult to identify and quantify using these techniques
26
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
(Raskin et al., 1994). As discussed in section 2.2.1, the use of molecular techniques may
hence aid in the investigation of anaerobic microbial growth under variable conditions.
2.3 The inhibition of LCFAs on anaerobic microoganism
With the increasing world population and rising living standards, disposal of plant oily
waste has become increasingly an issue. Industrial wastewater and domestic sewage have
different concentrations of lipids ranging from 20% to 65% (Quemeneur et al., 1994;
Broughton et al., 1998). Olive mill and palm mill wastewater, domestic kitchen wastewater
and other plant oily waste are the consequences of increasing populations. The chief
constituents of plant oil are long chain acids of varied colour.
Because of the close
similarities between triglyceride molecules found in different oils, the densities and
viscosities of most oils lie within a narrow band of higher viscosity and higher density oil.
Olive oil is a pale yellow or greenish-yellow liquid consisting primarily of oleic acid,
palmitic acid and linoleic acid. Palm oil is a yellow-brown, soft and buttery solid at room
temperature and is made up of triglycerides of oleic, palmitic and lauric acid.
In some olive growing countries of the Mediterranean area (Greece, Italy, Lebanon,
Portugal, Spain, Syria, Tunisia and Turkey), olive oil mill effluent production was reported
as more than 30 million m3 per year (Beccari et al., 1996). There are many difficulties
involved in the treatment of this product. Anaerobic digestion is being increasingly used to
treat these effluent products due to the process producing considerably less waste sludge
27
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
than aerobic processes as well as energy savings. Moreover, anaerobic digestion could
easily be restarted after several months of shut-down (Beccari et al., 1996). Biodegradation
of palm oil effluent and organic oil and fats from restaurants and other commercial kitchens
by anaerobic digestion to environmentally acceptable products has been reported (Ibrahim et
al., 1984; Stoll and Gupta, 1997).
Although animal fat waste is not as toxic as petroleum, the large quantities of animal oily
waste produced daily (such as wool scouring effluent and the fat waste of the meat industry),
constitute a problem for adequate disposal. There is currently no effective method available
to treat this waste. Animal fat is generally a dark brown and fixed oil with a distinct odour.
Its chief constituents are hydrocarbons, pyridine bases and amines.
Wool grease is
comprised of both oxidised and unoxidised grease, both of which have a melting point of
around 40 ºC. After refining, wool grease (called lanolin) forms a highly complex mixture
of lipids with high molecular weight. Some batch experiments performed on the anaerobic
digestion of wool wax and sheep tallow showed potential on wax hydrolysis and degradation
(Broughton et al., 1998; Gutierrez, 1999).
Treatment of wool scouring effluent by anaerobic bioflocculation (Lapsirikul et al., 1993)
indicated potential benefits. Research into the use of anaerobic digestion for petroleum
waste treatment (Kleerebezem et al., 1997) and anaerobic digestion of milk-fat waste
(Petruy and Lettinga, 1997), manure waste (Ahring et al., 1992), slaughterhouse waste
(Salminen et al., 2000) and cheese whey waste (Yilmazer Yenigun, 1999) have been trailed
with good results. In the past twenty years, with the development anaerobic digestion, more
28
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
emphasis has been placed on LCFA degradation due to lipids and LCFAs forming a major
part in domestic and industrial wastewater.
The effects of LCFAs on anaerobic microorganisms were first observed in rumen bacteria
(Demeyer and Henderrickx, 1967; Viviani, 1970) the focus being on digestive and
nutritional problems. Their effect on rumen bacteria was initially examined using poorly
defined mixed rumen culture and/or involved the use of in vivo experiments in which the
rumen bacteria could not be monitored.
Galbraith et al. (1973) found that low
concentrations of LCFAs could induce lysis of bacterial cell protoplasts at pH 7.4. A few
studies involved in vitro studies in which pure methanogen groups were used. It was
difficult however, to show the real inhibition of LCFAs on anaerobic digestion using this
method, furthermore, the toxicity of LCFA was also not investigated for each anaerobic
microbial group.
Hwu (1997) reported that LCFA toxicity varies with the type of anaerobic sludge and is
more correlated to their physical characteristics, specific surface area and size distribution.
He also suggested that oleic acid toxicity was higher in the flocculent sludge than in the
granular sludge under the three tested temperatures of 30, 40 and 55ºC. This was due to a
bigger specific surface area being available for LCFA absorption in flocculent sludge than in
granular sludge with a subsequently higher potential for damage of anaerobic bacteria.
Rinzema (1988) and Koster and Cramer (1987) also reported that inhibition of LCFA was
mainly related to its concentration and that the ratio of LCFA and biomass was less
important.
Most of the research of lipid anaerobic degradation involved chemical
29
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
engineering, mathematical modelling and reactor designing (Lettinga et al., 1997). The
detailed microbiological studies of this field were limited to the traditional culture-based
techniques, which could not adequately monitor the complex anaerobic microbial
communities present (Raskin et al., 1994b).
The inhibition of anaerobic digestion by LCFAs was investigated primarily in the past as its
effect on methanogens using the measurement of parameters such as VFA, methane volume
and difference of influent or effluent with these parameters (Koster and Cramer, 1986;
Hanaki, 1981). Methanogenesis was inhibited by concentrations higher than a specific
concentration of LCFAs. There was however, a slight stimulation of methanogenesis when
the concentration was lower than this concentration (Hanaki et al., 1983).
Numerous
interactions of complex metabolic groups have been observed to occur during anaerobic
digestion of LCFAs, such as saturation of unsaturated LCFAs (Hanaki et al., 1983), the
possible pathway for specific LCFA (Weng and Jeris, 1976).
Little is known about
subpopulation shifts under LCFA inhibition. Study of anaerobic microoganisms during
LCFA digestion is therefore necessary.
2.3.1
The physicochemical characteristic of LCFAs
Fatty acids normally contain even numbers of carbon atoms which are from 14 to 24 in
biological systems. Fatty acids with 12 carbon atoms or above are considered to be LCFAs.
LCFAs have varying chain lengths and degrees of saturation (Table 2.4).
Moreover,
different types of raw material will vary in their main constituents (Table 2.5). This results
in a range of LCFA ratios occurring in wastewater, depending upon the origin of the waste.
30
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
However, oleic (cis-9-octadecenoic) acid and palmitic (hexadecanoic) acid are generally the
most abundant unsaturated and saturated LCFAs present in wastewater.
Triacylglycerols are the most abundant lipids and can be hydrolysed to fatty acids and
glycerol as shown:
O

O
CH2—O—C—R1


R2—C— O —CH
O


CH2—O—C—R3
+ 3H2O
Triacylglycerol
CH2OH
Hydrolysis
HO—CH
CH2OH
Glycerol
+
O

R1—C—OO

R2—C— O- + 3H+
O

R3—C—OFatty acids
Fig.2.3 Pathway of triacylglycerols hydrolysis. R1, R2, R3 present alkyl groups in the
equation.
Approximately 93-96% of COD in lipids is conserved in LCFAs from the
hydrolysis of lipids.
31
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Common
Systematic
Carbon
Double bond
name
name
numbers number
Formula
Lauric
Dodecanoic
12
0
CH3(CH2)10COOH
Myristic
Tetradecanoic
14
0
CH3(CH1)12COOH
Palmtic
Hexadecanoic
16
0
CH2(CH2)14COOH
Palmitoleic cis-9- Hexadecenoic
16
1
CH3(CH2)5CH=CH(CH2)7COOH
Stearic
Octadecanoic
18
0
CH3(CH2)16COOH
Oleic
cis-9- Octadecenoic
18
1
CH3(CH2)7CH=CH(CH2)7COOH
Linoleic
cis-9,12-Octadecadienoic 18
2
CH3(CH2)4(CH=CHCH2)2(CH2)6COOH
Linolenic
cis-9,12,15-Octadecatric 18
3
CH3CH2(CH=CHCH2)3(CH2)6COOH
Arachidic
Eicosanoic
20
0
CH3(CH2)18COOH
Gadoleic
cis-9- Eicosenoic
20
1
CH3(CH2)9CH=CH(CH2)7COOH
Arachidonic cis-5,8,11,14-Eicosatetric 20
4
CH3(CH2)4(CH=CHCH2)4(CH2)2COOH
Behenic
Docosanoic
22
0
CH3(CH2)20COOH
Erucic
cis-13-Docosenoic
22
1
CH3(CH2)4(CH=CHCH2)4(CH2)2COOH
24
0
CH3(CH2)22COOH
Lignoceric Tetraacosanoic
Table 2.4 Different LCFAs in lipids (Hwu, 1998)
32
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Name
Lauric Myristic
Palm oil
1.4
Palmitic
Stearic
Arachidic
Palmitoleic
Oleic Linoleic
42.9
4.8
0.5
0.7
39.0
10.0
Olive oil
14.3
2.4
0.5
1.4
71.4
5.5
Linseed oil
7.1
3.3
15.2
11.4
57.1
Ground nut oil
16.2
2.4
0.8
41.0
32.9
1.0
Cotton seed oil 1.4
25.7
2.9
1.0
15.2
51.9
1.0
Soya bean oil
1.0
11.0
4.8
0.5
21.9
49.0
7.6
Beef tallow
1.0
2.6
28.1
20.0
0.5
3.8
37.6
2.9
Chicken fat
1.4
21.0
4.3
0.5
6.7
42.4
20.0
1.4
Lard
1.0
2.4
26.2
13.3
0.5
3.8
43.3
7.1
26.7
32.9
1.4
0.5
33.8
4.3
2.0
6.0
2
39
13
27.6
16.7
48.3
5.1
8.1
0.7
29.2
1.0
Cocoa butter
Whole milk
7.0
6.0
Raw sewage
16.4
5.2
0.9
30.5
Linolenic
0.5
0.5
Domestic
2.2
Table 2.5
LCFA composition in raw material and wastewater (weight percentage)
(Rinzema, 1988).
Although it is normally considered that LCFAs are degraded via β-oxidation both under
mesophilic and thermophilic conditions, ω-oxidation is also thought to be involved in LCFA
degradation (Jeris and McCarty, 1965). The pathway can be described as shown in Figure
2.4 (Hwu, 1997):
33
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Released during the hydrolysis of lipids, LCFAs can be degraded to acetate and hydrogen
via β-oxidation. During β-oxidation, unsaturated and saturated LCFAs are both activated by
the acyl-CoA synthetase present in the cytoplasmic membrane of cells. LCFAs enter into the
microbial cells before β-oxidation degradation starts (Fig.2.5). The transport of LCFAs into
the cytoplasm is a complex process (Ratledge, 1994). Several hypotheses based on the study
of the membrane-associated transport of E. coli (Nunn, 1986) were proposed. Nunn (1986)
suggested that LCFAs were adsorbed onto the outer membrane receptor or pore, then were
transferred across the cytoplasmic membrane to bind to acyl-CoA synthetase (Fig.2.5). The
molecules were then released into the cell after activation by acyl-CoA synthetase.
Saturated long-chain acyl-CoA then undergoes β-oxidation. Unsaturated molecules of longchain acyl-CoA require saturation before entering β-oxidation. During the β-oxidation
cycle, the long-chain acyl-CoA is broken down into a two-carbon fatty acetyl-CoA which
can be released as acetic acid without any further activation. The remaining shortened longchain acyl-CoA remains as substrate in the β-oxidation cycle until all the carbons are broken
down to acetyl-CoA, and finally to acetic acid.
34
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Unsaturated
LCFA
LCFA
Saturated
LCFA
activation
HS-CoA, ATP
HS-CoA, ATP
activation
Unsaturated longchain acyl-CoA
H2
saturation
FAD
HS-CoA
Saturated long-chain
acyl-CoA
FADH2
H 2O
acetyl-CoA
NAD+
NADH
Fig.2.4 The pathway of LCFAs degradation in β-oxidation. Adapted from Hwu (1997)
acylCoA
sythetase
FLP
LCFA
Long -chain
acyl-CoA
β-oxidation
PS
CoA+ATP
OM
PG
IM
Fig.2.5 Model of LCFA transformation (Nunn,1986). FLP: a possible receptor; OM: outer
membrane; PG: peptidoglycan; PS: periplasmic space; IM: inner membrane.
35
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Acyl-CoA synthetase is hence important in oxidative degradation, both in LCFA transfer
and in β-oxidation. Furthermore, the enzyme appeared to effectively detoxify the fatty acids
as the CoA ester was not inhibitory.
2.3.2 The toxicity of LCFA on anaerobic groups
LCFAs are well known inhibitors to various anaerobic microorganisms at millimolar
concentrations (Rinzema et al., 1994). It was suggested that the toxicity of LCFAs to
bacteria was a result of the absorption of LCFAs by cell walls or membranes, leading to the
damage of transport systems or protective functions (Galbraith and Miller, 1973).
In this
theory, it was thought that areas covered with LCFAs had altered permeability, resulting in
the lysis of protoplasts (Galbraith et al., 1971). This assumption was also demonstrated by
an experiment in which unsaturated LCFAs (ULCFA) had a higher inhibition level than
saturated LCFAs (SLCFA) (Koster and Crimer, 1986; Thies et al., 1994). Due to a bent
molecular shape arising from double bonds, unsaturated LCFAs have a greater cover area of
cell wall per molecule than saturated LCFAS.
Increasing the number of double bonds in
LCFAs was shown to increase inhibition in previous studies (Demeyer and Henderrickx
1967). Furthermore, the saturation degree (ratio of SLCFA / ULCFA) was reported as the
key factor governing membrane fluidity. Hwu (1997) indicated that microorganisms such as
thermophilic methanogens with cell membranes composed of a high proportion of saturated
LCFA, were more susceptible to LCFA toxicity than mesophilic methanogens, whose cell
membranes contained more unsaturated LCFA (ULCFA). It was also reported (Thies et al.,
1994) that the addition of the detergent Tween 80 (mainly comprised of oleic acid) resulted
36
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
in a change in the fatty acid composition of the thermophilic cell membrane to higher levels
of ULCFA with a subsequent fall in SLCFA. Kodicek and Worden (1945) found that the site
of action for toxicity of LCFAs was the wall of the bacterial cell with the effects of
unsaturated fatty acids rising with increasing unsaturation in Gram-positive bacteria. It was
also found that Gram-positive bacteria were more sensitive to the inhibition of LCFAs than
were gram-negative species (Nieman, 1954).
However, the actual mechanism of LCFA
toxicity is not known.
It was observed that addition of CaCl2 and other soluble calcium salts reduced the inhibition
of LCFAs due to precipitation of the salts with the LCFAs
(Hanaki et al., 1983).
Angelidaki et al. (1990) found that the toxicity of LCFAS could be decreased by the
addition of bentonite due to its capacity to flocculate LCFAs. However, these chemical and
clay materials were also toxic to anaerobic bacteria and/or were difficult to degrade.
Mousa and Forster (1998, 1999) found that some growth factors, such as glucose and
glycine behaved antagonistically with the inhibitory effects of gallic acid. Hwu (1997)
reported that using glucose at 35% in total COD weight enhanced the degradation of sodium
oleate. These results indicated that addition of the growth factor or other organic substrates
during LCFA degradation might decrease the toxicity of LCFA. However, the effects of
addition of LCFA on the growth of the microbial communities are poorly studied. During
anaerobic digestion, complex microbial communities are involved in the degradation
process. The nutrient requirements for specialized bacterial groups are different.
37
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Fermentative bacteria can hydrolyse organic compounds (protein, lipid and carbohydrates),
while syntrophic acetogens and methanogens degrade the fermentation products to methane
and carbon dioxide. The difference of composition of the nutrient content may affect the
growth of specific bacteria groups. Previous research reported that glucose and cysteine
were necessary to the growth of anaerobic bacteria (Balch et al., 1979), while the detailed
nutritional requirements during lipid treatment and the nutrient composition balance needed
during this process is still not known. Thus, a thorough investigation of the nutritional
requirements of anaerobes during lipid treatment using physico-chemical techniques
combined with molecular techniques will be of benefit to wastewater treatment industries.
2.3.3 The adaptation of methanogens to the inhibition of LCFA
It was reported that during the start-up of an anaerobic treatment system of lipid containing
wastewater, a lag phase in degradation of LCFAs would occur (Rinzema, 1988 ; Hwu,1997).
This appeared to be an adaptation phase due to an inhibition of LCFA degradation while
acetogenic bacteria capable of LCFA degradation grew in number. Reactors with lower
concentrations of LCFA feed normally exhibited faster adaptation periods than those fed
with higher concentrations (Rinzema, 1988; Hwu, 1997). This indicated that a higher
concentration of LCFA had a greater toxic effect.
Roy et al. (1986) found that both
methanogens and acetogens were greatly inhibited by LCFA.
A lag phase while
subpopulations of more LCFA resistant microorganisms grew could hence be expected.
Usually, the recovery of microorganisms in thermophilic digestion processes was faster than
38
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
in mesophilic treatments (Hwu, 1997). This might have been due to the faster growth rate
of those thermophiles which survived initial exposure to LCFAs.
Because the adaptation of acetogens and methanogens to LCFA had an unavoidable lag
phase, a two-phase treatment of LCFAs in which the acidogenic reactor was separated from
the methanogenic reactor was introduced. Phase separation was normally achieved by
keeping a short retention time in the acidogenic reactor to wash-out methanogenic bacteria
(Ghosh et al., 1975). In an acidogenic reactor, lipids are hydrolysed to LCFAs and then to
short chain fatty acids that are later used by methanogens in the methanogenic reactor.
Acidogens were shown to be less susceptible to LCFA inhibition and had a faster growth
rate than methanogens (Jeyaseelan McCarty, 1995). Some difficulties such as low to nil lipid
removal in acidogenic reactors have been reported however (Hanaki et al., 1987;
Fongastitkul and Mavinic, 1994). A thorough study of this field is therefore still required.
2.4 Anaerobic treatment of lipids containing wastewater
Oily waste treatment is a new and difficult field that has aroused great interest in the
environmental sciences.
Because of the LCFA toxicity described previously and the
propensity to form floating aggregates, lipids are difficult to degrade biologically. Although
some physical methods can be used prior to biological treatment, the use of chemical or
physical techniques to degrade lipid containing wastewater is very costly. For this reason,
more emphasis has been placed in recent years on improving the methods of biological
treatment.
39
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
The advantage to treatment of oily wastes by anaerobic digestion is that the method can be
exploited on a large scale as a simple and effective biotechnology process to reduce the
pollution caused by these wastes.
In theory, anaerobic digestion offers significant
advantages in energy terms. Aerobic waste treatments require an energy input and produce
end products of little value. Anaerobic digestion however, with its valuable main product,
methane, not only has the potential to treat wastes but also to reduce the energy usage of
petroleum based oils.
As a new technique then, anaerobic digestion of oily wastes is still facing some difficulties.
Nevertheless, with more intensive research using advancements in the fields of physicochemical technology and molecular biology, more progress for lipids treatment will be
achieved.
2.5 Difficulties and new approach of lipids anaerobic treatment
Anaerobic digestion has long been used for the stabilisation of wastewater sludges and is
now widely used in the treatment of lipid containing wastewater. Some of the reported
advantages of this process were:
1. Anaerobic digestion of lipid used readily available CO2 as an electron acceptor and had
no aeration costs (Lettinga et al., 1980; Beccari et al., 1996).
2. Anaerobic digestion produced lower amounts of sludge because the energy yields of
anaerobic bacteria were relatively low (Lettinga et al., 1980):.
40
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
3. The lipid waste are valuable sources of energy and fertiliser and most of the energy of this
process was found in the final product (CH 4 ) that could be used for power generation
(Salminien et al., 2000).
4. The process of lipid anaerobic digestion could be easily restarted after several months
shut-down and it was very useful for the seasonal oil production plants (Beccari et al.,
1996).
As a biological process, anaerobic digestion also had some disadvantages for lipid
degradation:
1. Difficulties experienced in anaerobic digestion of oily waste included the fact that lipid
compounds, as well as long chain fatty acids and phenolic compounds of the C-7 and C9 phenylpropanoic family, are difficult to degrade by microorganisms.
2. The high organic load was difficult to dilute (Beccari et al., 1996) and this caused a very
slow digestion process in oily waste anaerobic treatment.
3. Flotation or washout of granular sludge and fatty matter occurred at low waste loadings.
4. The hydrolysis products of lipid, LCFAs were shown to be highly toxic to anaerobic
miroorganisms.
These problems have only partly been overcome by the use of physical and chemical pretreatments, designing different engineering processes and modelling of digestion kinetics to
control the loading rate, HRT or other physico-chemical parameters. The activities of
microganisms in anaerobic digestion however, are difficult to control. Parameters that can
be controlled during digestion are pH, dissolved oxygen concentration (DO), oxidation-
41
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
reduction potential (ORP), load of organic compounds and nutrients, hydraulic retention
time (HRT) and sludge retention time (SRT).
The relation between these parameters and
the ecology of the microorganisms involved in the process is not clearly defined.
The general performance of anaerobic digestors and the diversity of wastes that can be
treated have been increasing steadily due to an array of breakthroughs related to reactor
design, operating conditions, and biocatalyst engineering.
After UASB, an expended
granular sludge bed (EGSB) and a staged multi-phase anaerobic (MPSA) reactor system
were rapidly developed.
The two-phase anaerobic digestion process has been widely used in organic wastes,
especially lipid containing wastewater (Beccari et al., 1996) but it still has problems in the
areas of poor supernatant quality and relatively low methane production (Fongastitkul et al.,
1994). However, the two-phase system is very attractive due to its prevention of lipid
inhibition of methanogens (Komastsu et al., 1991) and the enhancement of lipid hydrolysis
(Beccari et al., 1996).
One new approach reported was the use of granulation-enhancing additives (Lettinga et al.,
1997). Biomass retention through adequate granulation is of utmost importance in UASB
technology; firstly, in order to obtain a good effluent quality and secondly, to avoid the
wash-out of the slow-growing anaerobic bacteria. It was suggested that addition of synthetic
42
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
polymers could shorten the onset of sludge granulation after start-up of an anaerobic
sequencing batch reactor from four months to one month (El-mamouni et al., 1998). Water
absorbing polymer particles were also shown to enhance the formation and stability of the
granules (Imai et al., 1997).
Another new approach used anaerobic biomass to degrade toxicants present in wastewater,
soil and river sediments. This method used specific biomass with different residence times,
anaerobic tunnel reactors and sequential anaerobic-aerobic treatment to treat various
toxicants (Londry, 1997; Trevizo et al., 1999).
The major limitation to these new
applications was the long adaptation period (up to months or years) required before the
microbial community achieved maximal conversion rates of a new compound.
Nevertheless, the potential for using biotechnology and genetically engineered
microorganisms are apparent.
Recently, with the development of integrated municipal waste management, the number of
co-digestion plants has been increased due to the advantage of full-scale co-digestion of
organic waste (Kubler et al., 2000; Edelmann et al., 2000; Ahring et al., 1992). The lipids
added to anaerobic digester treating animal manure could be converted effectively (Ahring
et al., 1992). However, studies of anaerobic co-digestion of organic materials were mainly
in applying physico-chemical methods (Kubler et al., 2000). The understanding of the
microbial mechanisms for co-digestion of three organics is still unknown. Thus a study for
the microbial community shifts in co-digestion of lipids, carbohydrates and proteins to find
the optimal co-digestion conditions is necessary.
43
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
The efficiency of co-digestion might be related to the multiple sources of nutrients that serve
a similar physiological function. The term "mixed substrates" encompass all types of
nutrients; such as the carbon sources, energy sources, N-sources, etc. Mixed substrates are
very relevant in waste management sciences. The great majority of biological waste
treatment processes involve the metabolism of mixed substrates.
It was observed that the co-metabolism from hydrocarbon-utilizing bacteria which could
grow on methane as the sole source of carbon (Alexander, 1985). They were methanotrophs
(methane-oxidizing bacteria) and not able to utilise or metabolise higher alkanes such as
ethane or propane as sole carbon sources. When the bacteria were grown on a mixture of
methane, ethane and propane, the cells were found to not only utilise methane, but ethane
and propane also get oxidized to products such as acetaldehyde, acetic acid, propionic acid
and acetone, respectively. The investigation of biodegradation toluene and other toxic
pollutants in wastewater treatment showed that these toxicants could be degraded in the
presence of mixtures of organic compound which provide source of carbon and energy
(Stenstrom, et al., 1989). Alexander (1985) defined co-metabolism as a process in which a
toxic compound is mineralised or converted to non-hazardous metabolic products and it
does not provide energy or nutrient for microbial growth. With co-metabolism techniques, It
was investigated that pentachlorophenol (PCP) could be degraded by mixed microorganisms
in an anaerobic, fixed-film reactor and by supplementation with 1g/L glucose. Some
toxicants, such as polymers, such as poly (β-hydroxyalkanoates) (PHA) and two carbon
44
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
chlorinated aliphatics (eg. TCE) could be degraded by co-digestion with other substrates
(Alexander ,1985).
Although the process of anaerobic digestion has progressed considerably since its first
reported use, little is known about the microecosystems of anaerobic bacteria and the
microbial effects on wastewater treatment processes. Recent advances in this area include
the use of 16S rRNA hybridisation probes to describe natural communities of methanogens
(Amann et al., 1990; Raskin et al., 1995) and the investigation of microoganism structure
with antibodies (Conway and Macario, 1982). These approaches may have potential in
improving the current understanding of biological wastewater treatment. All the functional
behaviour of anaerobic digestion ultimately rests on the metabolic pathways of its
microorganisms and the microecosystem of bacteria. The speed, the end products and the
reaction rate of the anaerobic digestion process depend upon the growth of anaerobic
microbial populations.
Although syntrophic and competitive interaction in microbial
populations present in anaerobic decomposition is known to occur, the extent of knowledge
in this field is limited. This is a result of the limitations of traditional microbiological
techniques, such as selective enrichment, pure culture isolation and number counting. With
advancements in molecular techniques, in-depth understanding of the nutrient requirements
of microorganisms during anaerobic digestion will be achieved. From this knowledge, it is
envisaged that the physicochemical and engineering aspects of anaerobic digestion will
further improve.
45
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
2.6 Research objective
Current difficulties in lipid anaerobic digestion processes occur during the hydrolysis of
lipid. The inhibition of LCFA on methanogens and bacteria can lead to the shut down of
anaerobic reactors. The current status of knowledge of the microbiology of lipid anaerobic
digestion is mainly about the inhibition of LCFA and lipids on anaerobes. The study of the
reduction of the inhibition by mechanism such as substrate addition however, is not
exhaustive.
Therefore, this thesis has the following objectives:
•
To quantify the effect of substrates, such as glucose and cysteine on LCFA inhibition of
specific anaerobic microbial groups
•
To define the function of glucose and cysteine on LCFA inhibition on the granulation of
anaerobic bacteria
•
To clarify the influence of glucose and cysteine on the recoveries of microbial groups
under LCFA inhibition
•
To examine the effects of addition of proteins and carbohydrates on lipid hydrolysis
46
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
•
To explore the optimum substrate composition as well as other physicochemical
conditions for anaerobic co-digestion of lipid with carbohydrate and protein
•
To compare the efficiency of two-phase and one-phase UASB reactors on lipid
degradation when co-digested with protein and carbohydrate
2.7 Outline and outcomes of the thesis
Chapter 1 provides the general introduction and summary of the thesis.
Chapter 2 presents the current knowledge, background and motivation for the research. The
microbiology of lipid and LCFA anaerobic digestion as well as the current research on lipid
anaerobic digestion is discussed.
Chapter 3 describes the experimental methods and techniques used for the determination of
physico-chemical parameters and the microbiological investigation.
Chapter 4 characterises the role of typical nutrients, glucose and amino acid (cysteine) on the
reduction of toxicity of LCFA (sodium oleate) in batch experiments. The effects of nutrients
were examined by investigating of physicochemical parameters and by fluorescent in-situ
hybridisation (FISH).
The effects of LCFA and addition of glucose and cysteine co-substrates on granule formation in
lab-scale UASB reactors is presented in Chapter 5.
47
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 2: Literature Review
Investigations of the a recovery or adaptation of anaerobes to LCFA inhibition by adding
different substrates (glucose and cysteine) in lab-scale UASB reactors is discussed in
Chapter 6.
Chapter 7 compares the efficiency of monitoring the microbial population and physico-chemical
parameters to control the lipid anaerobic degradation process. The microbial population shifts
of acidogenic and methanogenic bacteria in one-phase reactors during lipid degradation were
monitoring by using microbiological and FISH technique.
The influence of protein and carbohydrate addition on lipid hydrolysis in the acidogenic and
methanogenic stages of a two-phase UASB reactor under different experimental conditions is
given in the Chapter 8.
The optimal ratio of lipid, protein and carbohydrate on the growth of methanogenic and
acidogenic bacteria in acidogenic and methanogenic reactors and the effect of the composition
of carbohydrate and protein on lipid degradation is discussed in the Chapter 9.
Chapter 10 presents the conclusions and recommendations.
48
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
Chapter 3
Materials and methods
This chapter summarises the general materials and methods used in this research. The
details of the specific methodology for each experiment can be also found in the related
chapters.
3.1 Experimental lab-scale reactors and set up
3.1.1 Anaerobic sludge and substrate feeding
The suspended sludge from the Woodman Point Wastewater Treatment Plant (Woodman
Point, WA, Australia) that mainly treated municipal wastewater was chosen as all inoculum
for the experiments described in Chapter 4 and 5, which investigated LCFA toxicity and
granulation, respectively. The sludge was stored at 40C in tightly closed plastic containers
and used without any preparation. The inocula for the experiments described in Chapters 6,
7, 8 and 9 was obtained from the hybrid UASB reactors of the Spearwood Waste Treatment
Plant (Spearwood, WA, Australia). The sludge of the Spearwood Wastewater Treatment
Plant was chosen for the acidogenic reactor because this plant was mainly fed with
wastewater from a slaughterhouse, rich in lipid and protein. The sludge would therefore
adapt more easily to the lipid feeding in the experiment.
In addition to sodium oleate or canola oil as feed (table 3.1), other substrates (table 3.2) were
used to feed the reactors for reactivation or start-up. Canola oil was emulsified by sonication
49
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
with an ultrasonic clearer (Branson, Sonifier 450) for 2 h (40% duty cycle and 7 output
control) before feeding The details of feed concentrations varied according to the
experimental requirement and can be found in separate chapters.
Substrates
Batch reactor One-phase UASB reactor
Two-phase UASB reactor
Canola oil
No
No
Yes
Sodium Oleate
Yes
Yes
No
Basal Medium
Yes
No
No
Yeast extract
No
Yes
Yes
Starch
No
Yes
Yes
Table 3.1 Substrates for different reactors
50
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
NAME
AMOUNT (g/l)
KH2PO4
0.4
K2HPO4
0.2
NH4Cl
0.3
KCl
0.5
NaCl
1
NaHCO 3
1.8
Na2CO3
0.5
CaCl2.2H2O
0.05
MgSO4.7H2O
0.1
Yeast Extract
1
Resazurin Indicator
1ml
Sodium Dithionite
until colourless
pH 7-7.5
Table 3.2. Chemical composition of basal medium (Lapsirikul, 1993).
3.1.2 Set up of UASB reactor and two-phase UASB reactor
The glass UASB reactor was operated with a water jacket by circulating water from a water
bath at a constant temperature as shown in Figures 3.1 and 3.2. Substrate was fed by
peristaltic pump and the feed tank was stirred continuously with the mixer. Biogas was
collected from the top of the reactor and led through 10% NaOH before being measured by
51
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
water displacement using an inverted cylinder and the gas flow rate was measured using a
downward displacement of acidified water (0.05M H2SO4). The effluent tube was U shaped
and maintained by a liquid seal to prevent gas escaping through the line. The two-phase
reactors used in the experiments were made from Perspex (polycarbonate) material. They
possessed the same shape and working volume (2.0 L). A water bath was employed to
maintain the water jacket temperature around 370C.
Fig. 3.1 Photograph of single and two-phase UASB reactors.
52
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
Biogas collector
5
NaOH (10%)
6
Pump
Effluent container
P
P
Water bath
Stirrer
3 3
Feed container
Reactor
A
53
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
Gas collector
Gas collector
Pump
Pump
P
P
Effluent
Feed container
H
pH adjust container
Acidogenic
reactor
Water bath
effluent
container
Methanogenic
reactor
B
Fig.3.2. Schematic diagram of experimental single-phase (A) and two-phase (B) UASB
reactors.
54
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
3.2. Batch experiment for LCFA toxicity test
Forty ml of the basal medium (Table 3.2) and 10 ml of the flocculent sludge were pipetted
into the serum bottle of 120 ml working volume.
The vial was flushed with N2 /CO2 (2:8
in v:v) before sealing with a silicon rubber cap and an aluminum clasp.
Subsequently, the reactors were placed in a 370 C water bath stirred at 30 strokes/min. Prior
addition of sodium oleate (SO), glucose and amino acid, the reactors were reactivated for 48
h to degrade the residual substrate. After reactivation, 20 ml of the experimental substrate
solution (according to the experiment in chapter 4) were injected into the reactors to get
different concentrations of different substrates.
The control was added with the same
volume demineralized water. The vials were incubated in the water bath for 48 h. All the
experiments were conducted in triplicate. The headspace of reactors was flushed again
before collection of the methane gas for GC. The methane composition of the reactors was
measuring using GC (see section 3.6.1). The samples for FISH experiments were gained
and fixed freshly for whole cell hybridization and DAPI stain quantification (see section 3.3
& 3.4).
55
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
3.3
Quantification
and
identification
of
anaerobic
bacteria
groups
by
oligonucleotide probe hybridization
3.3.1 Oligonucleotide probes
The four group-specific rhodamine-dyed 16S rRNA hybridisation probes, MB310,
MC1109, MG1200 and MSMX860, which target the communities of methanogens were
bought from Synthetic Genetics, California (U.S.A). They targeted all three orders of
known methanogens, Methanobacteriales, Methanococcales, Methanomicrobiales.
them, families Methanomicrobiaceae,
In
Metahosarcinaceae, Methacorpusculaceae,
Methanosarcinaceae made up the order Methanomicrobiales. The three rhodamine-dyed
oligonucleotide, ARC915 and EUB338, and SRB385 were purchased from SigmaAldrich Pty. Ltd. (Australia).
Probes EUB 338 and ARC 915 were used to investigate
the two domains Bacteria and Archaea, separately, while SRB was applied to target
Sulfate-reducing bacteria. The other fluorescein-isothiocynate (FITC)-dyed probe,
EUB338, was ordered from Synthetic Genetics, California (U.S.A) for granule location
analysis. The probes were 5’-end labelled with fluorescent dye and sequences for 7
probes were extracted from the previous reference (Table 3.4).
56
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
Probe Name
Sequence (5’-3’)
Target groups
Reference
1. MB310
GCAACATAGGGCACGGGTCT
Methanobacteriaceae
Raskin et al. 1994a
2. MC1109
GCAACATAGGGCACGGGTCT
Methanoococcales
Raskin et al., 1994a
3. MG1200
CGCATAATTCGGGGCATGCTG
Methanomicrobiaceae
Raskin et al., 1994a
Methanocorpusculacea
Methanoplanacea
4. MSMX860 GGCTCGCTTCACGGCTTCCCT
Methanosarcinaceae
Raskin et al., 1994a
5. SRB385
CGGCGTCGCTGCGTCAGG
Sulfate-reducing bacteria
Amann et al., 1990b
6. EUB338
GTCGCCTCCCGTAGGAGT
Bacteria
Amann et al., 1990a
7. ARC915
TCGCACCGTGGCCGACACCFAC
Archaea
Stahl and Amann, 1991
Table 3.3 Oligonucleotide probes used in this study
3.3.2 Cell fixation and whole–cell hybridization
About 10 ml of granule and suspended sludge were taken from the reactors and promptly
washed by phosphate-buffered saline (PBS ) solution (130 mM sodium chloride, 10 mM
sodium phosphate buffer) (Raskin et al.,1994 a & b). After washing, the granule sludge
was disintegrated by an ultrasonic clearer (Branson, Sonifier 450) with continuously
microscopy inspection to monitor disaggregation and possible cellular disruption.
One ml cell suspension was obtained and centrifuged at 13,000 rpm for 4 min. Half of
the supernatant was removed and the cell pellets were resuspended in the same volume of
57
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
fixative solution (described below) or ethanol (99.5%) depending on the probes used.
The suspended cells were incubated at 40C for about 18h for fixation.
The fixative solution was prepared with 4% Paraformaldehyde in 200mM sodium
phosphate buffers (pH 7.2) and not more than 8h before use. After centrifugation 4 min
at 6000 rpm, the pellets were washed by centrifugation with 9 volume of PBS (130mM
NaCl, 10 mM sodium phosphate) and 1 volume Igepal CA-630. The supernatant was
removed and cells were resuspended in 50% ethanol and stored at –200C for several
weeks without significant decrease of rRNA content.
Four µl well suspended samples were spread on the well of six-well Teflon-coated slides
purchased from Proscitech Australia.
After air drying, the slides were dehydrated by
three successive immersions of slide in 50, 80 and 100% ethanol solution for 2 min each.
They were air dried again and the slide prepared for hybridisation. Ten µl hybridisation
and 1 µl rhodamine-labelled oligonucleotide probe (50-80 ng) were added in the wells
and mixed. The slides were incubated in a 42 0C moisture chamber for 2 h, rinsed twice
with the hybridisation solution and washed 15 min at 42 oC.
They were washed three
times with double distilled water at room temperature and air dried in the dark at room
temperature.
58
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
3.3.3 Quantification of different microbial groups
The hybridised cells were air dried and viewed immediately under oil immersion objective
with a UV light on a Leitz DIAPLAN 307-148 microscope fitted for epifluoresence excited
in the green region of the spectrum. Ten fields were counted in each well to determine the
number of cells fluorescing in the rhodamine emission region. Black and white Kodak TriX Pan 400 film was used for photomicrographs. Exposure times were between 40 s - 6 min
for epifluorescence.
3.4 Total cell counts with DAPI method
The fixation of cell was the same as described above. The fixed cells were immobilized on
the hybridisation well and stained by DAPI (6-diamidino-2-phenylindole) solution for 7 min
by adding 20 µl of 40 µg/ml DAPI solution (30% formamide, 0.9% Nacl, 20mM Tris HCl at
pH 7.2, 0.1% SDS) to the hybridisation well. The cells were washed with the same solution
without DAPI. Air dried and viewed at Leitz DIAPLAN 307-148 microscope fitted for
epifluoresence excited in the UV region of the spectrum. The counting was performed at 10
different microscopic fields (Amman et al., 1990b)
59
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
3.5 Investigation of granular structure with scanning and transmission electron
microscopy.
Scanning electron microscopy (Philips 501B) and transmission electron microscopy (Philips
CM 100) from the Veterinary School, Murdoch University, Western Australia were used for
the analysis. The methods of fixation granule samples were modified from Macleod et al.
(1990) and Zhu and Gu (1997).
The sludge samples collected from the reactors were gently washed with a 0.1 M phosphate
buffer solution and fixed with 2.5% glutaraldehyde in sealed 50 ml serum bottles overnight
at 4oC. The fixed samples were washed twice for 10 min, each in 0.1 M cacodylate buffer.
The granules were fast frozen in liquid nitrogen with aluminium block and cleaved with a
motar and pestle. The granules were dehydrated through graded series ethanol (30%, 50%,
75%, 90%, 100%) for 15 min each. Dehydrated samples were air dried in a desiccator and
mounted on a stub and sputter coated with gold. Samples were viewed with a SEM (Philips
501B) at 10 kv.
For TEM , the granular samples were washed with 0.1 M phosphate buffer solution twice
gently and quickly fixed with 3.5% glutaraldehyde for 24 h in 50 ml serum bottle which
sealed with butyl rubber stoppers and aluminium seals. Treated with 0.1 M phosphate
buffer and fixed in 1% OsO4 for 1.5 h, they were washed twice with buffer then dehydrated
in a series of 30%, 50%, 75%, 90%, 100% w/w acetone-water solution for 15 min each.
After being washed twice with propylene oxide over 10 min. and left in epoxy resin
60
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
overnight on a rotator, they were embedded in capsules by using fresh epoxy resin and
baked at 60-65 0C for 24h. The hardened embedded sludge was cut with a glass knife. Thin
sections (silver or golden colour) were chosen and placed on copper grids. Finally, these
sections were stained first with uranyl acetate then with lead citrate, and dehydrated in 80%,
90%, absolute alcohol. They were critical-point dried and gold coated, and viewed with
TEM.
3.6 Confocal laser scanning microscopy (CLSM).
The confocal laser scanning microscopy (Olympus, Fluoview, BX50) from the State
Agricultural Biotechnology Centre, WA, Australia was used to view the florescence of
rhodamine and FITC from different probes.
3.6.1 Fixation and sectioning of granule
The method for the granule sectioning was modified from Chui and Fang (1994) and
Sekiguchi et al. (1999). A sample of 50-60 granules was washed three times with 0.1 M
phosphate buffer solution at pH 7.2 and then fixed in a phosphate buffer solution with 10%
formadehyde at 4 0C overnight. It was then dehydrated with 50%, 80%, 100% ethanol
solution twice at 15 min each, and then immersed in a graded series of xylene/ethanol
solution, in which the xylene concentration gradually increased from 50% to 100%.
Granules saturated with xylene were immersed in molten paraffin 50% (v/v) for 30 min, and
then immersed in 100% paraffin twice. Five to ten granules were transferred to the 10 ml
vial and immersed in molten paraffin at 60 – 65 0C overnight, allowing enough paraffin to
61
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
penetrate into the granules. Three to four granules paraffin–embedded granules were put
into the peel-off mold and cooled to the room temperature. The thin paraffin section were
sliced by a microtome (Leitz Model 1400) and mounted on gelatine-coated glass slide. The
slide was air-dried, then put back in a 650 C oven for half an hour before dewaxing with
xylene.
3.6.2
Hybridisation and CLSM scanning investigation
For only rhodamine dyed probes, 2 µl (25-50 ng/µl) rhodamine labelled oligonucleotide
probe and 10 µl hybridisation buffer (0.9 M Nacl, 20 mM Tris-HCl (pH 7.2) 0.01% sodium
dodecyl sulfate) were added to each well. Then hybridisation of the 16S rRNA-targeted
oligonucleotide probes to granule section were performed at 40 0C for 16 hours. After
hybridisation, the granule section was washed 4 times with hybridisation buffer, air dried
and observed by CLSM.
For double staining hybridisation, FITC labelled probe was hybridised and washed first, and
rhodamine labelled probe was followed for the second hybridisation. After the second
hybridisation, the granule sections were washed with the hybridisation buffer for 4 times, air
dried and observed by CLSM (Sekiguchi, et al., 1999).
62
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
3.7 Analysis of gas and effluent of batch and UASB
3.7.1 Gas chromatography (GC)
The methane in gas was analysed by GC (Varian Model 3700) using the following
parameters:
Column ID and length:
1.8 m &2 mm
Column temperature:
1000 C
Carrier Gas:
Nitrogen
Injector temperature:
2000 C
Sample volume:
1 ml
Detector:
FID
Detector temperature:
2000 C
The volatile fatty acid (VFA) in samples was prepared by centrifuging at 15, 000rpm for 10
min to remove the suspended solids (Hwu, 1998). The supernatants were acidified by the
addition of 1% HCl to convert the VFAs into the free acid form. Then the sample was ready
for VFA examination with the same method as methane gas analysis, except for the column
temperature (200 0C) and sample volume (10 µl). The standard curve of acetate, propionate,
butyrate were made with different concentrations, respectively.
63
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
The long-chain-fatty acids were determined by GC (SHIMAZU, GC 17A) with a column
AT-225 (Altech, Perth, Australia) using the following parameters:
Column ID:
0.25 mm
Column Length:
30 metres
Film thickness:
0.25 µm
Detector:
FID
Detector temperature:
2750C
Injector temperature:
2500C
Oven temperature:
1250C
Carrier gas:
Helium
Flow rate:
1.2 ml/min
Pressure:
12 psi
Split ratio:
100:1
Sample size:
1.0 µl
Solvent:
Hexane
The standard LCFA methyl esters, palmitic acid methyl esters, oleic acid methyl esters,
liolenic acid methyl esters and linolenic acid methyl esters were bought from Sigma Aldrich
Pty., Ltd. (NSW, Australia).
64
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
3.7.2 High performance liquid chromatography (HPLC)
Samples (1 ml) were acidified with 4 drops of 10 N HCl and extracted with 10 ml petroleum
ether (Lapsirikul, 1993). A standard concentration of SO was also analysed in parallel.
After 1 hour mixing, the ether phase was transferred to another tube and entirely evaporated
in 800 C water bath. Then HCl and dry methanol (1:19 v/v) were added and mixed and put
back to the 800C water bath for 1 h. After esterification, 2 ml of petroleum ether were added
and vortexed for 5 min.
One ml of the ether phase were added into a glass vial and capped
for analysis.
The sodium oleate in sample was analysed with HPLC (Varian 2010), equipped with Varian
2050 UV Detector and Varian 4290 Integrator. HPLC was performed with sperical 5 µm
LiChrospher-NH2 column packing. The effluent A and B were mixtures of nhexane/methanol. The UV detector wavelength was 217 nm.
3.8 Measurements of lipids, carbohydrates and proteins and methods for
measurements of COD, total suspended solids (TTS), volatile solids (VS), suspended
solids (SS) and volatile suspended solids (VSS).
All the measurements of samples were applied in duplicate or triplicate.
3.8.1
Lipid analysis
65
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
The concentration of lipid was determined gravimetrically after extraction of lipid from the
samples by petroleum ether, according to the Soxlett extraction method (APHA, 1985).
3.8.2
Carbohydrate analysis
The anthrone method was applied for the analysis of carbohydrate as described by
Raunkjaer et al. (1994). The mechanism of this method is based on the colour reaction
between carbohydrate and the anthrone reagent. This method has a high specificity for
carbohydrate and the colour can be measured accurately by spectrophotometer.
The anthrone reagent was a 0.1% solution of anthrone made up in 75% (V/V) sulphuric acid.
It should be prepared freshly each day of analysis. Solution of glucose of 5, 20, 40, 80 and
100 mg/l was prepared as standards for a standard curve. One ml samples or standards were
pipetted into Pyrex test tubes. Then 1ml of cold anthrone reagent was added and the test
tubes were closed and shaken. The test tubes were put into the 1000C water bath for 20 min
and then were placed to the 40C refrigerator for 10 min. Finally, the samples were measured
spectrophotometer at 625 nm.
3.8.3
Protein analysis
The Lowry method was applied in the experiments because of its sensitivity to low
concentrations of protein (Raunkjaer et al., 1994). Bovine serum albumin (BSA) was used
as standard. The principle behind the Lowry method of determining protein concentrations
66
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
lies with the reactivity of the peptide nitrogen[s] with the copper [II] ions under alkaline
conditions
and
the
subsequent
reduction
of
the
Folin-Ciocalteay
phosphomolybdicphosphotungstic acid to heteropolymolybdenum blue by the coppercatalysed oxidation of aromatic acids.
Reagents preparation : A. 2% Na2CO3 in 0.1 N NaOH; B. 1% NaK Tartrate in H2O;
C. 0.5% CuSO4.5 H2O in H2O; D. 48 mL of A, 1 mL of B, 1 mL C, 1ml
E. Phenol Reagent - 1 part Folin-Phenol [2 N]: 1 part water
To obtain the standard curve, eleven sets of three 16 x 150 mm test tubes in triplicates were
set up in rack. 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 µl of BSA was added to these tubes.
Two ml of solution D was then added to each test tube. Incubate at room temperature for 30
minutes before addition of 0.2 mL of dilute Folin-phenol solution (E). Each tube was
vortexed immediately and incubated at room temperature for 30 minutes. Determination of
the absorbance of each sample at 600 nm was plotted against amount of protein (mg) to
obtain standard curve.
For the analysis of samples, set up triplicate assays as described for standard curve. The
data from the standard curve were used to determine the concentration of unknown samples.
3.8.4
Other analysis
COD, total suspended solids (TTS), volatile solids (VS), suspended solids (SS) and volatile
suspended solids (VSS) were analysed as described in the Standard Methods (APHA, 1985).
67
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter3 Materials and Methods
68
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by
addition of co-substrates in batch tests
4.1. Introduction
Long-chain fatty acids (LCFA) are products of lipid hydrolysis in anaerobic wastewater
treatment. They are mostly degraded into acetic acid and hydrogen by acetogens via the ßoxidation pathway and finally converted to methane by methanogens. Due to the toxicity of
LCFA on anaerobic microorganisms, both acetogens and methanogens suffer greatly from
inhibition of LCFA in anaerobic digesters (Hwu et al., 1996). It was reported that LCFA
inhibition process was so significant that the methanogenic activity was lost only after
several minutes of exposure (Koster, 1987). The reason for inhibition by LCFA could be
related to the adsorption of LCFA onto the cell membrane resulting in the damage of its
transport and protective function and finally in the lysis of the cell (Demeyer and
Henderrickx, 1967). It also might be related to the fact that LCFAs change the membrane
permeability and affect the surface tension, thus inhibiting the division of the cell (Rinzema,
1988).
The toxicity of LCFA on methanogenic bacteria has been positively correlated to the
concentration of LCFA (Hanaki, 1981). Severe inhibition can be caused even at 3 to 5 mM
concentration of LCFA (Hanaki, 1981; Hwu et al., 1998). The ratio between LCFA and
biomass appears to be of less significance for the toxicity of LCFA (Koster and Cramer,
68
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
1986; Rinzema, 1988). Besides the concentration, previous investigations demonstrated that
the toxicity of LCFA was also related to the temperature, pH and origin of the sludge
(Galbraith and Miller, 1973; Hwu, 1997). Koster (1987) reported that the addition of
calcium chloride decreased the inhibition of LCFA by forming the insoluble calcium salt of
LCFA and thus reducing the concentration of LCFA. However, this was only a short-term
strategy, as the toxicity of LCFA would reappear after several hours. Angelidaki et al.
(1990) also found that the addition of bentonite (a clay mineral) reduced lipid inhibition and
caused a significant increase in the degradation of lipid. The possible reason for this effect
might be the flocculation capacity of bentonite with the lipid. However, it seemed that the
bentonite could not be degraded and had no benefit to the bacterial population. Although
the simple addition of calcium or bentonite could decrease the toxicity of LCFA in pilot
studies, its application to full scale lipid treatment has not been yet reported.
This chapter discusses investigations on the effect of addition of co-substrates on the
inhibitory effects of LCFA during anaerobic digestion from physico-chemical and microbial
aspects. In the experiments reported here, sodium oleate (SO) was selected as the model
compound for LCFA because it is the most abundant of all LCFAs in wastewater (Hwu,
1997). Previous research has reported that glucose and cysteine were necessary for the
growth of anaerobic bacteria (Balch et al., 1979). Therefore glucose and cysteine were
chosen as the co-substrates for digestion with LCFA. In addition, glucose and cysteine are
likely to be produced as intermediates during the digestion of complex substrates containing
carbohydrates and protein. The effects of glucose and cysteine as well as their different
concentrations and composition on toxicity of sodium oleate were also compared. The
69
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
counteraction of glucose and cysteine to sodium oleate toxicity was investigated by
checking the production and composition of gas and VFA. The effects of sodium oleate,
glucose and cysteine on the shift of specialised anaerobic bacterial population were also
monitored by 16S rRNA fluorescence in-situ hybridisation (FISH) technology and linked to
the functional changes of batch reactor. All investigations were carried out as batch assays
in serum vials.
4.2. Material and Methods
The preparation of inoculum and its reactivation were described in 3.2. The methods of
quantification and identification of anaerobic bacteria groups can be found in 3.4. The
methane production was examined with the method shown in 3.6.1.
4.2.1. Examination of individual effect of glucose, cysteine and sodium oleate on
methanogenic activity
The individual effect of glucose, cysteine and sodium oleate on the anaerobic process was
investigated by feeding different concentrations of these substrates separately.
concentrations of glucose and cysteine were chosen as 0, 0.5, 1, 2, 3, 4
The
g/l, and sodium
oleate as 0, 1, 2.5, 5, 10, 20 mg/l. The source of sludge was a suspended sludge obtained
from the Woodman Point Wastewater Treatment Plant (WA, Australia). The methods for
batch assays were as described in section 3.2. The initial concentrations of glucose, cysteine
and SO were set in the different vials with addition of 20 ml substrate solution. A control
assay was set up with the same volume of demineralized water.
70
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
After 24 h of incubation, the content of methane in the gas phase of the vials was examined
by GC as described in 3.7.1, and the methane content was used as an index of methanogenic
activity. The loss of maximum methanogenic activity in each reactor was expressed as a
percentage of the uninhibited maximum methanogenic activity of control.
All the
experiments were conducted in triplicate and the results quoted as mean values.
4.2.2. Comparison of toxicity concentration of sodium oleate with the addition of
glucose and cysteine
The concentrations of SO were adjusted to 0, 1, 2.5, 5, 10, 20 mg/l by injecting 20 ml SO
solution into the vials. The vials were fed with the optimal concentration (determined in
4.2.1) of glucose, cysteine or glucose and cysteine together, respectively. In the vial, which
contained both glucose and cysteine, the amount of each was half the optimal concentration.
The control reactors were only fed with the same concentration of glucose, cysteine or
glucose and cysteine together, respectively without SO. Then the headspace of the vials was
flushed with N2 / CO2 and return back to the incubator.
The content of methane gas in the gas phase was determined after 24h of incubation. The
relative methanogenic activity was expressed as the percentage of methane volume
compared to the control. All the experiments were conducted in triplicate and the results
quoted as mean values.
71
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
4.2.3. Effect of the addition of glucose and cysteine on the degradation of SO in batch
tests
This experiment was performed in four groups. The concentration of SO was chosen higher
than the concentration that inhibited the methanognesis determined in 4.2.1. The adsorption
of SO during the experiment was checked separately by adding SO to autoclaved sludge.
While one group was taken as a control with fed SO, three groups were fed glucose, cysteine
or both glucose and cysteine. The concentration of glucose and cysteine were chosen as 0,
0.5, 1, 2, 3, 4 g/l, respectively. The total concentration when both glucose and cysteine were
added was 0, 0.5, 1, 2, 3, 4 g/l with cysteine and glucose being added at 1:1 (w/w) ratio.
The degradation of SO was calculated by subtracting the SO residuals in the batch reactor
after 24 h incubation and the adsorbed quantity of SO of sludge from the total amount of SO
added into the vial. SO was measured following the methods outlined in 3.7.2.
4.2.4 Effect of glucose and cysteine on degradation rate of SO
The experiment was performed in four groups. The control group was prepared by only
feeding SO (5 mg/l). The other three groups contained glucose (2 g/l), cysteine (2 g/l) or
both glucose (1 g/l) and cysteine (1 g/l) in addition to SO at a concentration of 5 g/l. The
concentration of SO was examined after 5, 10, 15, 20, 25 and 30 h of incubation. SO was
determined following the methods described in 3.7.2.
72
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
4.2.5. Growth of specialized microbial groups under different substrate composition
The microbial communities were analysed after 30 h by using 16S rRNA probes. Two
domain specific probes, ARC 915 (targeting domain Archaea) and EUB 338 (targeting
domain Bacteria) were used to investigate the Archaea and Bacteria population which
covered all the microorganisms present in the batch reactors. Because the sulfate existing in
cysteine could benefit the growth of Sulfate-reducing Bacteria (SRB), the group specific
probe SRB 385 (targeting SRB) was used to monitor SRB and evaluate the competition
between methanogen and SRB. To study the microbial shifts in domain Archaea, order
Methanobacteriales (targeted by MB310)) and four families Methanosarcinaceae (targeted
by MSMX860), Methanomicrobiaceae, Methanocorpusculaceae and Methanoplanceae
(targeted by MG 1200) were analysed separately. All the probes listed above circumscribed
most of the currently known microbial groups which are involved in LCFA, glucose and
cysteine anaerobic degradation. The competition and coexistence among Archaea, Bacteria
and SRB was observed by comparing the shifts of the total cell number examined by wholecell hybridization methods as described in 3.3. All the experiments were conducted in
triplicate and the error bars indicated the standard deviations.
4.3. Results and Discussion
4.3.1. Effect of co-digestion of glucose and cysteine on the toxicity of sodium oleate
Figure 4.1a presents the methanogenic activities of the vials containing different
concentrations of SO and SO along with glucose and/or cysteine (Fig.4.1b). The inhibitive
73
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
concentration (IC) of SO which caused 50% methanogenic activity lost was defined as IC50
in the experiment with SO concentration higher than 1mg/l, the methanogenic activity
sharply decreased and IC50 was about 2.5 mg/l.
In previous research, the reported
concentration of LCFA which caused CI50 inhibition on methanogenesis was different.
Hanaki et al. (1983) reported that 5 mg/l loading of SO resulted in 50% inhibition for
methanogenesis, while it was 130 times higher in Koster and Cramer (1987) experiment. In
Hwu’s (1997) experiment, the IC50 was about 4 mg/l. While Koster and Cramer (1987)
speculated that this difference might be due to different culture medium applied in the
sepecific experiment. Hwu (1997) suggested that it could also be related to the sludge
originals and temperature.
As would be expected, the methanogenic activity increased as the concentration was
increased up to 1 g/l for cysteine and 2 g/l for glucose. As the concentration of glucose and
cysteine was increased beyond these, the methanogenic activity dropped. At 3 g/l and 4 g/l
of glucose or cysteine, the methanogenic activity was only 40% of the activity observed for
the control. Even though it would be expected that methanogenic activity increased with the
addition of more substrate, at high concentration of substrate, it is expected that the
methanogenic capacity of the inocula would have been exceeded, which in turn, caused the
VFA to accumulate resulting in a drop in pH. Low pH would inhibit methanogenic activity.
Hence it was shown that for the basal medium and sludge used in the assays here, 1 g. L -1,
of cysteine and 2 g. L -1of glucose was optimal.
74
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
a
100
90
SO
80
70
60
50
40
30
20
10
0
0
2
4
6
8
10
12
14
16
18
20
SO concentration (mg/l)
b
160
Glucose
140
Cysteine
120
100
80
60
40
20
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Substrate concentration (g/l)
Fig.4.1.
Methanogenic activity from feeding glucose, cysteine and SO at different
concentrations.
The toxicity of SO on methanogenic activity and the effects of the addition of glucose and
cysteine on this toxicity is shown in Fig.4.2. The IC50 concentration of SO was about 2.5
mg/l and was very similar to that observed in the batch tests discussed previously (Fig.
4.1a).
The addition of glucose increased the IC50 of SO from 2.5 mg/l to about 12 mg/l,
75
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
and the addition of cysteine increased the IC50 of SO to about 5 mg/l. Addition of both
glucose and cysteine increased the IC50 of SO to 20 mg/l. This suggested that addition of
other organic substrate such as glucose and/or cysteine could relieve the toxicity of SO on
methanogenic activity. In terms of extent of enhancement, the addition of glucose and
cysteine was most effective in three feeding modes. Following this, glucose gave better
results than the addition of cysteine alone.
100
SO
90
SO+Glu
80
SO+Cys
70
SO+Glu+Cys
60
50
40
30
20
10
0
0
2
4
6
8
10
12
14
16
18
20
SO concentration
Fig.4. 2. Methanogenic activity when SO is co-digested with glucose (2 g/l), cysteine (1 g/l),
and glucose (1 g/l) together with cysteine (0.5 g/l).
The IC50 concentration of SO on the methanogenic activity was about 2.5 mg/l (Fig.4.1a).
In order to investigate the effects of glucose and cysteine on the degradation of SO, the
concentration of SO was chosen as 5 mg/l which was higher than the IC50 concentration of
SO.
76
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
Fig.4.3 shows that the amount of SO degraded increased with the addition of glucose,
cysteine or combination of glucose and cysteine. The results indicated that glucose had a
better effect than cysteine, while the combination of glucose and cysteine had the greatest
effect on enhancing the degradation of SO. It is in accordance with the result of Figure 4.2,
which also suggested that the combination of glucose and cysteine had the optimal effect on
decreasing the toxicity of SO. The result also showed that the addition of 1 g/l glucose and
1 g/l cysteine had the best effect on increasing the degradation of SO. Increasing the
concentration of glucose and cysteine beyond these values did not enhance SO degradation,
but on the contrary, the residual SO concentration increased. As explained before, at
concentration above 2 g/l, it is likely that VFA accumulation could cause pH to drop, which
in turn inhibited methanogenic activity, when glucose together with cysteine was added.
5
4.5
4
3.5
3
2.5
2
1.5
1
Glu
cys
Glu+Cys
0.5
0
0
1
2
3
4
Substrate concentration (g/l)
Fig.4.3. The effects of glucose and cysteine on the degradation of SO. The concentrations of
glucose and cysteine in Glu+Cys group were 1:1 in w:w.
77
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
Figure 4.4 shows the effect of glucose, cysteine or glucose and cysteine on the rate of
degradation of SO. In the vials which were fed only SO, very little of SO was degraded
over the 30 h of the test. This is in agreement with results of batch assays discussed in
Fig.4.1a in which SO inhibited the methanogenic activity severely. The addition of glucose
and cysteine increased the degradability and degradation rate of the SO. The combined
addition of glucose and cysteine increased more significantly the degradation rate of SO. By
supplementing with glucose and cysteine, about 80% of the initial SO was degraded in 25
hours. In addition, more SO was degraded over the 30 h. Glucose alone increased the
degradation rate higher compared to the addition of cysteine alone. This is in accordance
with the results shown in Fig.4.2 in which glucose was more effective than cysteine in
decreasing the toxicity of SO.
SO
SO+Glu
SO+Cys
SO+Glu+Cys
6
5
4
3
2
1
0
0
5
10
15
20
Test time (hours)
25
30
35
Fig.4.4. Effect of glucose (2 g/l), cysteine (2 g/l) and combination of glucose (1 g/l) and
cysteine (1 g/l) on the degradability of SO (5 mg/l).
78
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
The experiments above demonstrated that the addition of limited concentration of glucose
and cysteine significantly decreased the toxicity of SO on methanogenic activity and
enhanced the degradation of SO. The results also showed that the combination of glucose
and cysteine was most effective in reducing the inhibition by SO, compared to the separate
addition of glucose and cysteine. Glucose has been reported as a growth factor which
decreased the inhibition of some compounds both in aerobic and anaerobic digestion (
Mousa and Forster, 1999). Beccari et al. (1996) found that oleic acid (3g COD/l) could not
be degraded without the supplementation of glucose. Hwu (1997) observed that use of
glucose and acetate as co-substrates enhanced the degradation of LCFA mixture in
anaerobic reactor. The role of the glucose was that of an exergonic substrate which provided
the energy for the microbial populations. As a source of nitrogen, cysteine is also well
known to benefit the acidogenesis, acetogenesis and methanogenesis in anaerobic digestion.
However, high concentration about 90 mg N/l also inhibited the anaerobic process (Samso0n et al., 1991).
4.3.2. Population shifts during degradation of SO with glucose, cysteine or glucose and
cysteine addition.
The microbial community structure analysis carried out in the previous section (Fig. 4.4) at
30 h are shown in Fig.4.5a. The control sample was the original inoculum diluted by
addition of demineralised water, the volume of which was the same as the volume of feed
used in the other tests. After feeding with 5 mg/l SO, the number of cells of Archaea,
Bacteria and SRB all decreased. Among, the decrease of Archaea was the most significant.
79
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
This might indicate a higher sensitivity of Archaea bacteria to SO compared to Bacteria
cells and SRB cell. The addition of glucose (2 g/l) alone with the SO significantly increased
the numbers of SRB, Archaea and Bacteria cells compared to the addition of SO only and
this might be linked to the enhanced methanogenic activity and SO degradation discussed in
Fig.4.2. When adding cysteine (2 g/l), the increase in population of Archaea and Bacteria
cells was not as high as that observed with the addition of glucose. However, the increase of
SRB was more significant than that in glucose addition. The presence of sulphur in cysteine
might stimulate the growth of SRB. When adding glucose (1 g/l) and cysteine (1 g/l)
together, the population of Archaea, Bacteria and SRB all increased and recovered numbers
similar to the control level. Therefore, the co-digestion of SO, glucose and cysteine was
most beneficial to the growth of Archaea, Bacteria and SRB.
Figure 4.5b shows the shifts within the Archaea community observed for the different feeds
after 30 h. The members of Methanococcales (MC1109) were not detected in the inoculum
for the different feeds. Methanobacteriales (MB310) were the most abundant group with
SO feeding alone, suggesting that Methanobacteriales were the least inhibited by the
toxicity of SO. In contrast, Methanosarcinaceae (MSMX860) were the least abundant,
indicating that Methanosarcinaceae was the most sensitive to the inhibition of SO. When
fed with glucose, cysteine or glucose and cysteine, the percentage of Methanosarcinaceae
increased to the highest value in Archaea, suggesting that the three feeding modes all
benefited the growth of Methanosarcinaceae. When fed glucose and cysteine, the structure
of the three methanogenic groups was similar to the control, implying that the
80
a
16
ARC915
14
12
SRB385
10
EUB338
8
6
4
2
SO+Glucose+cysteine
SO+cysteine
SO+Glucose
b
MG1200
MSMX860
SO+glucose+cysteine
MB310
SO+cysteine
MC1109
SO+glucose
SO
90
80
70
60
50
40
30
20
10
0
control
Methanogen population (%)
SO
0
Control
Cell density (LogC/ml)
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
Fig. 4.5. Microbial population shifts for different feeds (a), and methanogenic community
structure in batch reactors (b).
Data are mean values of triplicate tests and error bars
indicate the standard deviations.
81
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
feeding of glucose and cysteine promoted the optimal recovery of Archaea communities.
This might also be linked to the fact that the mixture of glucose and cysteine had the best
effect on the reduction the toxicity of SO.
The microbial investigation presented in this study suggested that glucose and cysteine
enhanced the degradation of LCFA in anaerobic digestion through the recovery of microbial
communities. Addition of glucose increased the growth of sub-populations of the microbial
communities, especially Archaea cells in anaerobic reactor. The reason might be related to
the fact that it provided the necessary growth factors for Archaea microbial communities,
and hence helped the Archaea microbial communities to survive under LCFA toxicity or
recover from the inhibition of LCFA. Addition of cysteine was not as effective as glucose in
increasing the number of Archaea cells but it was more effective in stimulating the growth
of SRB. In recovery, even the SRB population in cysteine feeding was higher than glucose
or glucose and cysteine feeding, the recovery of methanogenesis and SO degradation were
lower there, implying that methanogenesis and SO degradation was more linked to the
recovery of Archaea cells. Use of cysteine in the experiment provided some information
related to the effect of addition of amino acids in LCFA degradation. However, further
investigation on different amino acids is required. The addition of a mixture of glucose and
cysteine was most effective in enhancing the growth of Archaea cells. It optimally increased
the growth of the Methanosarcinaceae family which is the most important group involved
in the methanogenesis (Raskin et al., 1990b). Glucose and amino acids are required in
anaerobic digestion as they are important nutrients for the growth of anaerobic
microorganisms (Balch et al., 1979). The results suggested that the addition of glucose and
82
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
cysteine relieved the toxicity of LCFA by stimulating the growth of microbial communities,
which in turn, lead to more degradation of LCFA.
Mousa and Forster (1999) reported that the addition of the growth factor glucose (10 and 20
mg/l) reversed the toxicity of gallic acid. To date, some experiments have also demonstrated
that the glucose and other organic substrates could reduce the inhibition of some toxic
material in sludges (Hwu, 1997; Mousa and Forster, 1999). However, this mechanism is not
clear. It was speculated that some non-growth substrates or toxicant which do not assist in
cell division could be transformed by using the energy or enzymes produced during
degradation
of
other
co-substrates
or
growth
substrates
(http://www.uoguelph.ca/~hlee/418chap6.htm). The growth substrates, such as glucose or
amino acids could provide energy to support microbial growth and produce catabolic
enzymes which catalyse toxicant transformation.
During LCFA anaerobic digestion, the
availability of nutrients for anaerobe growth is more important due to the toxicity of LCFA.
The optimal ratio among glucose, amino acids and LCFA is very complex due to the
changing nutrient requirement as the community shifts. The results of the comparison of
microbial community shifts from this experiment showed that only specific concentrations
of glucose and cysteine could effectively decrease the toxicity of LCFA on microorganisms.
Too high or low concentrations could not decrease the toxicity of LCFA effectively.
It has been shown in industrial situation that co-substrates, such as carbohydrate, protein and
lipid could benefit the anaerobic degradation (Ahring et al., 1992). This might be related to
the fact that carbohydrate and protein could be hydrolysed to glucose and amino acids and
these components could benefit the growth of microorganisms under the inhibition of LCFA
83
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 4 Reduction of the toxicity of LCFA on anaerobic bacteria by addition of co-substrates in batch tests
which hydrolysed from lipids. The study of the effects of co-substrates on lipid anaerobic
digestion digestion will be performed in Chapters 7, 8, and 9.
4.4 Conclusion
In this chapter, the effects of glucose and cysteine on decreasing the toxicity of sodium
oleate and enhancing of its degradability were investigated in batch experiments. The
results showed that the addition of glucose and cysteine could reduce the toxicity of SO and
enhance the degradation of LCFA. The combination of glucose and cysteine was most
effective in reducing the inhibition of LCFA and stimulating the degradation of LCFA.
The microbial studies illustrated that SO seriously inhibited the growth of methanogens and
that the addition of co-substrates, glucose and cysteine decreased the inhibition of SO on
methanogens. While glucose was more effective than cysteine on stimulating the growth of
Archaea cells, cysteine benefited the growth of SRB cells. The co-digestion of glucose and
cysteine optimally stimulated the growth of Archaea, Bacteria and SRB cells and the
recovery of the methanogenic microbial groups to the structure prior SO inhibition. The
methanogenesis and SO degradation were more likely linked to the population of Archaea
cells than SRB cells.
84
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
Chapter 5 Influence of co-substrate supplementation on granule
characteristics in a LCFA fed single stage anaerobic digester
5.1 Introduction
High-rate anaerobic methanogenic systems are becoming increasingly popular to treat
wastewater from various industries including dairy, meat processing and beverage
industries.
Among the various designs that are commercially available the Upflow
Anaerobic Sludge Blanket (UASB) design is the most popular (Fang et al, 1994a). The
UASB design relies on the granulation of bacterial biomass with enhanced settling
properties to enable increased retention of the microbial consortia when compared to
continuous stirred tank digesters. The formation of granules is a complex process involving
inorganic and organic compounds as well as the bacterial communities.
The microstructure of UASB granules depends on physio-chemical parameters and the feed
composition (Fang et al., 1994a, b; Alpenhaar et al., 1994).
With TEM and SEM, it was
observed that granule microstructures differed depending on the substrate feed. Granules
(Fang et al., 1994a) from UASB reactor that was fed with sucrose as the major carbon
source was shown to be dense containing diverse bacterial population arranged similar to the
three-layered structure proposed by MacLeod et al. (1990). On the other hand granules fed
hydrolysed proteins had a coarsely packed structure with multiple cracks with satisfactory
settling characteristics (Fang et al., 1994b) and were predominantly smaller than the
85
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
granules fed carbohydrates. Data on the nature of microbial aggregation when fed with
lipids or LCFA are scarce. It is known that LCFA and lipids inhibit the formation of
granules (Sam-soon et al., 1991; Rinzema, 1988; Koster and Cramer, 1987) and that the
adsorption of LCFA on to the granules can result in its flotation and washout (Hwu, 1997).
Sam-soon et al. (1991) found that a digester fed with oleate could not produce granular
sludge and that the degradation of oleate was very poor. The difficulty encountered during
lipid and LCFA wastewater treatment has been mainly attributed to the high concentration
of these compounds in the wastewater (Beccari, 1996). Hwu (1997) reported that the
addition of glucose was likely to stimulate the degradation of LCFA.
Numerous studies have been reported on the microbiology of immobilized anaerobic
bacteria due to their important role in the structure and maintenance of granules (Wu, 1996;
Zhu et al., 1996). Using traditional microscopic methods including scanning transmission
electron microscopy (SEM) and transmission electron microscopy (TEM), several
investigators have found that not only methanogens were involved in granulation, but also
other bacteria, such as syntrophic bacteria were present in microcolonies (Macleod et al.,
1990; Wu, 1996). However, these techniques were not sufficiently specific to differentiate
between groups of microorganisms. In recent years, the bacterial distribution within granules
has been examined using whole cell in-situ hybridization with molecular 16S rRNA probes
(Sekiguchi et al., 1999; Harmsen et al., 1996). Sekiguchi et al. (1999) used 16S rRNA
probe in situ hybridization combined with confocal laser scanning microscopy (CLSM) to
investigate the spatial distribution of microbes in the granules.
86
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
A combination of TEM, CLSM and FISH techniques were applied to investigate the
characteristics of microbial agglomerates fed with LCFA in lab-scale UASB digesters. The
effects of glucose and cysteine addition on the degradation of LCFA and formation of
granules were also investigated in this study.
5.2 Materials and Methods
5.2.1 Anaerobic digesters and inoculum.
Reactors
Constituents
COD ratio
Total concentration
(gCOD.L-1.day-1)
R1
Oleate
R2
Glucose, Oleate
1:1
2
R3
Cysteine, Oleate
1:1
2
1:1:1
2
R4
Glucose, Oleate, Cysteine
2
Table 5.1 The composition of feeding nutrients
Four, 3 litre, mesophilic (37 0 C) lab-scale water jacketed anaerobic digesters, with diameter
95 mm and height 1100 mm, were used in the studies reported here. These were seeded
with sludge from a full-scale anaerobic sludge digester (35-370C) at Woodman Point
Sewage Treatment Plant (Woodman point, WA, Australia). The inoculum was stored at 4
0
C in tightly closed plastic containers for 5 days before use. The digesters were seeded with
this sludge and activated by feeding yeast extract (YE) and acetate (1:1 w/w at 1 g COD l1
.d-1) for 19 days until a stable volume of methane gas was achieved per day following
87
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
which each was fed with different substrates as shown in Table 5.1. The pH of the reactors
was maintained between 6.9 and 7.2 with the addition of HCl or NaOH (10 M). Feed into
the digesters was introduced from below at an upflow velocity of 0.027m/h. The
characteristics of the microbial agglomerates from the digesters were examined 160, 250 and
420 days after start-up.
5.2.2 Performance of reactors.
The performance of different reactors was monitored by measuring the methane gas
production rate and total COD removal rate.
The methane content of the gas was
determined using a Varian Model 3700 gas chromatograph fitted with a thermal
conductivity detector. The gas flow rate was measured using a downward displacement of
acidified water (0.05M H2SO4). The COD removal rate was calculated by measuring COD
of the effluent and in the feed using standard methods (APHA, 1992).
5.2.3 Fluorescence in Situ Hybridization (FISH) of section of granule.
Oligonucleotide probes were used to target most of the microbial populations in the digester.
The probes used and the corresponding targeted microorganisms are listed in Table 5.2.
Because these oligonucleotide probes had been widely used and demonstrated (Raskin et al.,
1994a; Sekiguchi, et al., 1999), the application using reference strain of bacteria was not
done. The method used for sectioning granule was as described by Chui and Fang (1994),
and the FISH method was modified from the techniques of Sekiguchi et al. (1999).
88
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
Hybridization of the 16S rRNA-targed oligonucleotide probes in the granule section was
performed at 400C for 3 h instead of 10h as described by Sekiguchi et al. (1999). This is
because too long a hybridization caused difficulties in washing off unhybridised
fluorescence probes. Sekiguchi et al. (1999) used Cy-5 and rhodamine labels to show the
green and red colour in the different layers. While the distribution of specific groups was
clearly visible, the intensity of the fluorescence signal from different groups could not be
compared easily. In this study, we chose to use only rhodamine labelled probes to be able
to compare the fluorescence intensity of different microbial subpopulations. In order to
investigate the syntrophic relation between Archaea members and Bacteria, section of
granules were double stained simultaneously by using FITC dyed EUB338 and rhodamine
dyed ARC915, MB310, MG1200 and MSMX860.
Probe Name
Targeted groups
Probe dye
ARC915
Archaea (domain)
Rhodamine
EUB338
Eubacteria (domain)
Rhodamine, FITC
MB310
Methanobacteriales (order)
Rhodamine
MG1200
Methanomicrobiaceae(family)
Rhodamine
Methanocorpusculaceae (family)
Methanoplanaceae (family)
MSMX860
Methanosarcinaceae (family)
Rhodamine
Table 5.2 Oligonucleotide probes used in the experiment.
89
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
5.2.4 Granule Characterization.
The whole granule structure was visualised through light microscope (LM). Granule size
was presented by diameter of the granule. TEM and CLSM methods and their sample
preparation were conducted as described by Fang et al. (1994a) and Sekiguchi et al. (1999).
While TEM provided information on the ultrastructure of bacteria, CLSM illustrated the
detailed location of specific bacteria in the granule.
5.3 Results and Discussion
5.3.1 Performance of the digesters.
Fig. 5.1 shows that the digester fed with only oleate achieved about 10% COD removal even
after 420 days operation, suggesting that the adaptation of anaerobic bacteria to the toxicity
of oleate is very difficult. This is in agreement with other observations (Koster and Cramer,
1986, Rinzema, et al., 1988). The digester fed with oleate and cysteine also showed a lower
adaptation process in which only around 50% COD removal was achieved after 420 days.
The COD removal of digester fed with glucose and oleate achieved 90% after only 50 days
of adaptation, whereas the digester fed with glucose, cysteine and oleate reached nearly 90%
of COD removal after 400 days, suggesting that glucose was more effective than cysteine in
reducing toxicity of oleate or in enhancing the activities of reactor. Although cysteine has
been shown to be a necessary nutrient in anaerobic bacteria and tissue culture, it has also
been reported that feeding cysteine might immediately reduce the generation of pelletised
mass (Sam-soon et al., 1991). These contradiction results might be related to the
concentration of cysteine, as high concentration of cysteine could inhibit the
methanogeniesis as shown in Figure 4.1 (Chapter 4).
90
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
100
90
R2
80
70
60
50
R3
R4
40
30
R1
20
10
0
4500
4000
3500
3000
R2
2500
2000
1500
1000
R4
500
0
R3
R1
Fig.5.1. The performance of the four UASB digesters. (a) COD removal efficiency, (b)
volumetric methane production.
91
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
a
b
c
d
e
f
g
h
i
92
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
j
k
l
m
n
o
p
q
r
s
t
u
93
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
A
B
C
D
E
F
G
H
I
J
K
L
94
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
Figure 5. 2
Comparison of granular structure of sludge from the different UASB
reactors by using LM (a, b and c), TEM (d, e and f) and CLSM (g to u showed the
sections hybridised with rhodamine labelled probes; A to L shows sections double
hybridised with rhodamine labelled methanogenic probes) and FITC labelled Bacteria
probes). Left column shows granule from R3 which was fed with oleate and glucose,
middle column shows granule from R2 which was fed with glucose and cysteine, and
right column shows granule from R4 which was fed with glucose, cysteine and oleate.
Distribution of microbial groups *
Feed
composit
ion
Oleate +
Glucose
Size
1-3mm
(a)
Oleate +
Cysteine
0-1mm
(b)
Oleate +
Cysteine
+
Glucose
1-3mm
(c)
Structure
Layered
structure
(a)
No layered
structure (b)
No layered
structure.
Fine outer
region (c).
ARC915
EUB318
MG1200
MSMX
860
MB310
Whole
Granule
(g)
Whole
Granule
(h)
Outer
region (j)
Whole
Granule
(m)
Sparsely
distribut
ed (n)
Whole
Granule
(p)
Outer
layer
(q)
Whole
granule
(s)
Sparsely
distribut
ed (t)
Mainly
outer
region
(0)
Whole
granule
(r)
Outer
region
(u)
Whole
Granule (i)
Mainly
outer &
centre
region (k)
Whole
granule
(l)
Table 5. 1. Comparison of structure and composition of granules from the digesters. The
respective figures from Fig.5.2 from which the observation were made are also indicated in
the table.
95
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
5.3.2 Comparison of microstructure of granules from digesters.
The digester fed with oleate did not form any ‘typical’ granules even after 420 days.
However, small flocs about 20-50µm were seen. This might be related to the inhibition by
oleate on the growth of microorganisms, and subsequent wash out of the microorgnisms
before proper granulation (Yan and Allen, 1997). It is well known that granulation depends
on several parameters including the wastewater composition and process conditions
(Schmidt and Ahring, 1996). Moreover, the bacteriological composition and structure of
granules adapted to different feeds are different (Grotenhuis et al., 1991). In a granular
sludge blanket, three distinct zones of behaviour can be identified. They include a lower
active zone, a middle active zone and an upper inactive zone (Wentzel et al., 1994). Samsoon et al. (1991) found that the pelletization and granulation did not develop in the sludge
bed when a typical LCFA, (oleate) was used as sole substrate to a digester. The results from
this study confirmed the previous observations. It might also imply that the failure of
granule development with the use of LCFA as the sole substrate is related to the difficultly
of LCFA degradation.
The characteristics and the distribution of microbial groups within the granules from the
digesters fed with oleate, and cysteine and/or glucose are given in Table 5.1. The
corresponding figures from which these observations were made are shown in Figure 5.2.
96
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
A granular sludge bed in the digester fed with oleate and glucose (R2) was established after
160 days of operation and a sample of the granules was taken after 250 days. They had a
clear-layered structure, which was very similar to the granules from a UASB treating
brewery wastewater (Fang et al., 1994a). The diameter of the granule was about 1-3 mm
and had a typical three-layered structure (Fig. 5.2a). The outer layer was loose, the middle
and inner layers were dense. The section observed by TEM illustrated that rod shaped
microorganisms and cocci microorganisms were densely packed together (Fig. 5.2d).
CLSM showed that the members of Archaea were distributed all over the granule (Fig. 5.2f).
These observations were very different from those of Sekiguchi et al. (1999) and Harmsen et
al. (1996) in which Archaea cells were mainly present in the inner and middle layers.
Observations showed that the members of Bacteria were mainly occupying the loosely
packed outer layer (Fig. 5.2j). This was in accordance with the investigation of Sekiguchi et
al. (1999) and Harmsen et al. (1996) in which Bacteria cells were mainly occupied in outer
layers. This is also in agreement with the structure proposed by MacLeod et al. (1990), in
which acidogenic bacteria were mainly in the outer layer of granule. The members of
families Methanomicrobiaceae, Methanocorpuslaceae and Methanoplanaceae (Fig. 5.2m),
and family Methanosarcinaceae (Fig.5.2p) and order Methanobacteriaceae (Fig.5.2s) were
observed in the whole granule (Fig. 5.2s). In three-layered structure model of MacLeod et
al. (1990), methanogens were in middle and outer layers, however, granules from this study
showed that methanogenic groups were distributed in the whole granule. The reason might
be related to the fact that the presence of oleate decreased the methanogen growth rate
resulting in high concentrations of acetate in whole granule which may have benefited the
growth of methanogen throughout the granule.
97
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
A granular sludge bed in the digester fed with oleate and cysteine (R3) developed after about
250 days of operation. The granules were very small (diameter 500 µm-1 mm). The
granules were very loose and lacked layered structure (Fig. 5.2b), and they were very similar
to the structure of glutamate-degrading granule structure discussed by Fang et al. (1994b).
The granule had a fine outer layer. The centre appeared to have large and thick clumps,
which was interspersed with groups of organisms of various morphologies showed by TEM
(Fig. 5.2e). There was no evidence of syntrophic microcolonies. There were large numbers
of bacterial cells of abnormal morphologies throughout the section, which might have been
non-methanogenic bacillus. CLSM images showed that Archaea cells were the dominant
cells in the granules (Fig.5.2h). The members of Bacteria (Fig. 5.2k) were located in outer
and centre region. The members of families Methanomicrobiaceae, Methanocorpuslaceae
and Methanoplanaceae (Fig. 5.2n), and order Methanobacteriaceae (Fig.5.2t) were
observed in the whole granule. The cells of family Methanosarcinaceae (Fig.5.2q) mainly
occupied in outer region of granules.
In the digester fed with oleate, cysteine and glucose (R4), a granular sludge bed was formed
after about 120 days of operation and granules were taken at day 250 with the same time
granules were sampled from the other two digesters. These granules were about 1-3mm in
diameter and larger and denser than the granules from the digester fed with oleate and
cysteine. The overall structure of these granules (Fig.5.2c) was similar to that of the granules
fed with oleate and cysteine (Fig. 5.2b) and did not exhibit the typical layered structure. The
granular structure of combined feeding of glucose, cysteine and oleate was very similar to
that showed in Batstone’s thesis and the reason for the lack of layered structure could be
explained by the slow growth of acidogens, acetogens and methanogens (Batstone, 2000).
98
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
In this experiment, there was no layered structure in the granules from reactors fed with
cysteine, implying that cysteine inhibited the formation of layered structure. TEM (Fig.
5.2f) indicated that the bacteria distributed in the granules were more complex and looser
than the granule from digester fed by oleate and glucose (Fig. 5.2d). CLSM images showed
that Archaea cells (Fig. 5.2i) existed in the whole space of the granules. The Bacteria cells
grew throughout the whole granule as different colonies (Fig. 5.2l). The spatial distribution
of
the
members
of
families
Methanomicrobiaceae,
Methanocorpuslaceae
and
Methanoplanaceae (Fig. 5.2o), and family Methanosarcinaceae (Fig.5.2r) were observed in
the whole granule. The cells of order Methanobacteriaceae (Fig.5.2u) mainly occupied in
the outer region of the granules.
The comparison of the rhodamine fluorescence intensity signals from three type granules
showed that the signal intensity generated by Archaea probes were stronger than the
intensity from Bacteria probe, suggesting that the growth rate of methanogens was the
higest. The Archaea probe signal generated by in the granules fed with oleate and glucose
or oleate, glucose and cysteine were stronger than that from the granule fed oleate and
cysteine, suggesting that the growth rate of Archaea was lower without glucose feeding.
This is in agreement in the activities of the reactors, in which the methanogenic activities
were the lowest in reactor fed with oleate and cysteine. In the granules from the three
digesters, Methanosarcinacea cells (targeted by MSMX860) had the strongest signals
compared to the other two methanogenic probes, suggesting that Methanosarcinace cells
were the dominant methanogens in three digesters.
99
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
With the double staining red rhodamine labelled methanogenic probe and green FITC
labelled Eubacteria probe, the syntrophic relation of Archaea members and Bacteria cells
was visualized as the yellow colour in the CLSM images (Fig. 5.2 A to L). The spatial
distribution of syntrophic groups were mainly located in the outer regions of granules,
implying that the outer region was more active than the inner region of granule.
It has been documented that the addition of glucose could enhance the biodegradation of
LCFA (Hwu, 1997). However, the effects of co-substrate addition on the granulation in
digesters fed with LCFA were still unknown. Results of this study showed that the
addition of co-substrates could not only decrease the inhibition by LCFA but also promote
granulation. However, the extent of granulation depended on the choice of co-substrates.
The rate for granulation was dependent on the feed composition. The study showed that
the granulation time in digester fed with oleate, glucose and cysteine was the shortest
among the four digesters. Granulation in the digester fed with oleate and glucose was
faster than granulation in the digester fed with oleate and cysteine. The reason for this
might be that glucose stimulated the growth of different bacterial groups and hence
enhanced the rate of granulation. Glucose feed has been found to be excellent for the
formation of the pellet or granules in UASB (Sam-Soon et al., 1991). Cysteine also
improved granulation but concentrations above 12.2mg/l could reduce the specific pellet
yield (Sam-Soon et al., 1987). Findings from this study indicated that supplementation
with cysteine alone did not enhance granulation but addition of glucose was necessary to
promote granulation.
100
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 5 Influence of co-substrate supplementation on granule characteristics in a LCFA fed single stage
anaerobic digester
5.4 Conclusion
Results indicated that addition of glucose and cysteine along with oleate not only
decreased the inhibition of oleate on methanogenic activity but also improved the
formation of granular biomass. The presence of glucose and cysteine enhanced the
granulation and methanogenic activity of biomass. However, addition of glucose alone
established higher COD removal and methane production rate much more rapidly than
cysteine alone or in combination, suggesting that glucose addition may have had the best
result in relieving toxicity of oleate. Microscopic images also showed that biomass from
the digester fed oleate and glucose formed the biggest granules (2-3mm), which in turn
may have had better settling properties, and closely resembled granules from a sucrose
fed digester (Fang et al., 1994a). As reported by Fang et al. (1994b) the size of granules
from digesters which were fed cysteine were smaller than that which was fed glucose but
addition of glucose in the presence of cysteine increased granule size. The CLSM images
of hybridised granule sections showed distribution of microbial groups was dependent on
feed compositions. ARC 915, which targeted Archaea cells including all methanogens
showed that methanogens within the granules from digester fed with oleate and cysteine
were less abundant than within the other two digesters. This is in agreement with the
methane production rate measurements from three digesters.
101
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition
by feeding glucose and cysteine in a single stage UASB reactor
6.1 Introduction
Previous studies (Hanaki et al., 1983; Hanaki 1981) and results in chapters 4 and 5 have
shown that even low concentrations of LCFAs can inhibit the methanogenic process. It has
been suggested that LCFAs are toxic for bacteria because of their damaging effect on the
transport channel of bacterial cell by inducing lysis of protoplasts (Galbraith et al., 1971,
Galbraith and Miller, 1973).
It has been observed that after a period of inhibition,
methanogenic activity would reappear suggesting that there was an adaptation of the bacteria
to the presence of LCFAs (Angelidaki et al., 1990; McAvoy et al., 1996). However, this
adaptation is a prolonged process and could last several months even under low
concentrations of LCFAs and was far too long for practical purpose. Furthermore, the only
way to recover the inhibited reactor in real plant is to re-inoculate although this is an
expensive operation (Rinzema, 1988; Koster and Cramer, 1986). Rinzema (1988) suggested
that the adaptation to LCFAs or the recovery of methane production under LCFA exposure
could be explained by the re-growth of a small number of surviving bacteria. It has also
been shown that the addition of calcium or bentonite can decrease LCFA toxicity as well as
enhance its degradation (Koster, 1987). However, the addition of calcium ion or bentonite
would only temporarily decrease the concentration of LCFA and the toxicity due to LCFA
would appear after several hours (Koster, 1987). These methods may therefore not be
102
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
practical.
Despite these observations, anaerobic digestion has been successfully used to treat complex
wastewater containing mixture of lipids, carbohydrates and proteins (Batstone, 1999). The
presence of co-substrates like carbohydrates and protein could reduce the toxicity of lipids,
or the hydrolysis products of carbohydrate and protein, i.e. glucose and amino acids could
reduce inhibition of LCFA. While Rinzema (1988) suggested that the recovery of bacteria
could not be related to the substrate composition, Mousa and Forster (1999) reported that
addition of glucose decreased the inhibition of methanogenesis due to gallic acid (C6H2
(OH) 3CO2H), a crystalline organic acid. The mechanism of counteraction of glucose on the
inhibition of gallic acid and the effect of glucose on anaerobic bacteria re-growth under
inhibition of LCFA might be related to the co-metabolism discussed at section of 2.3.2
(Chapter 2).
As previously described in Chapter 4, direct microbial investigation of microbial groups
under LCFA inhibition is scarce. Knowledge of re-growth of different microbial groups
during LCFA adaptation or recovery may assist to investigate the influence of glucose and
amino acids on the recovery process of anaerobic digestion under LCFA inhibition.
The aim of this study was to investigate the effects of addition of substrates on the reduction
of toxicity and recovery of anaerobic digestion from LCFA inhibition. FISH techniques
were applied to identify and quantify the shifts of main microbial groups. Sodium oleate
(SO) was used as the model LCFA in this study. It was used because of its well-understood
103
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
biochemical degradation pathway, its predominance in wastewater and its high solubility as
a sodium salt (Hanaki et al., 1983; Hwu et al., 1996). Glucose and cysteine were used as the
co-substrates. Glucose represents the hydrolysis product of carbohydrate degradation and
cysteine represents the product of protein degradation.
6.2 Materials and methods
6.2.1 Anaerobic digesters. Four mesophilic (37
0
C) water jacketed bench-scale UASB
reactors (labeled A, B, C, D) of 95 mm diameter, 1100 mm height and 2 L working volume,
were seeded with granular sludge as described in 3.1.2. The sludge originated from a hybrid
UASB reactor (35-370C) of Spearwood Wastewater Treatment Plant treating wastewater
from pig abattoir (Perth, WA, Australia). The sludge was stored at 4
0
C in tightly closed
plastic containers for 14 days before being used. The four digesters were reactivated for 19
days by continuously feeding a solution of 1.5 g COD. L-1 yeast extract and 1.5 g COD. L-1
acetate at a loading rate of 3 g COD. L-1.d-1 until stable methanogenic activity was attained.
After reactivation, the digesters A, B, C and D were fed only with a solution of 100 mg
COD.L-1 SO at loading rate of 100 mg COD. L-1.day-1 for 14 days after all digesters lost
almost all methanogenic activity i.e. no methane gas was produced. Digester A was treated
as the control for the investigation of the adaptation process by continuing to be fed with
SO. While the digester B was only fed with glucose 3 g COD.L-1, the digester C was fed
with only cysteine 3 g COD. l-1 and digester D was fed with both glucose and cysteine, 1.5 g
COD.L-1 of each . The loading rate of reactors B, C and D were all kept at 3 g COD. L-1 day1
. During the experiment, the pH of the reactors was kept between 6.9 and 7.2 with the
addition of drops of HCl or NaOH (10M).
104
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
6.2.2 FISH techniques.
The techniques of cell fixation, whole cell hybridization, quantification of microbial groups
under different condition and obtaining of Epifluorescence micrographs were as described in
3.3 and 3.4.
6.3. Results and Discussion
6.3.1. Comparison of performance of UASB reactors.
The data in Figure 6.1 illustrates the changes in methanogenic activities during the
experiment. After 10 days of reactivation, the production of methane gas from four reactors
was similar and about 900 ml. L-1.day-1. On day 19 onwards, reactors were fed SO,
consequently the methane production from the four reactors began to drop sharply.
Methane production (ml/l.day)
900
feed
SO
800
feed
glucose
cysteine
700
600
500
400
300
Reactor A
200
Reactor B
100
Reactor D
Reactor C
0
0
10
20
30
40
50
60
70
80
90
100
110
Time (days)
Fig. 6.1 The methanogenic activities after feeding different substrates
After 14 days of feeding with SO, no methane gas was produced from the four reactors.
105
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
From day 33, the feed to three out of four reactors was changed. Reactor A was controlled to
be fed with SO and was the control reactor. It did not produce any methane gas during the
whole experiment, showing that it was difficult for anaerobic bacteria to adapt or recover
from the inhibition of LCFA if LCFA was provided as the only carbon source. Methane
production from Reactor B began to increase within 2 days of switching to glucose feed. It
increased steadily and produced about 600 ml methane per day (about 65% of daily methane
gas production rate prior to inhibition by SO) 10 days after switching to glucose feed, and it
was maintained around 600 to 650 ml for the following 70 days. It suggested that glucose
could effectively stimulate the adaptation or recovery of anaerobic bacteria from the
inhibition of LCFA.
Compared to glucose, cysteine (Reactor C) had less effect on stimulating the methanogenic
activities. After 20 days, there was about 20% of daily methane gas production rate prior to
inhibition by SO recovered, and only about 30% of daily methane gas production was
recovered after 80 days.
When fed with glucose and cysteine (Reactor D), in the first 10 days, the methane gas
production was not as high as feeding only with glucose. However, the methane gas
producing increased sharply after 10 days and leveled off around 900 ml per day after 30
days. The feeding of combination of glucose and cysteine was not as efficient as only
glucose only for short recovery period from SO inhibition. However, it was more efficient
than feeding glucose only for long term recovery after SO inhibition. Even though the total
COD in different feeding was the same, the methane production were different in different
106
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
reactors, indicating that methanogenesis during the recovery of reactors was related to the
composition of organic substrates.
The degradation of SO in the control was kept very low (Fig.6.2), however, it increased
when fed with glucose, cysteine or glucose and cysteine, suggesting that glucose and
cysteine enhanced the degradation of SO.
In them, glucose was more effective than
cysteine in enhancing the degradation of SO, while the combination of glucose and cysteine
had the best result on the SO degradation.
Residual of Sodium Oleate (mg/l)
25
glucose
20
cysteine
15
control
glu+cys
10
5
0
33
40
47
54
61
68
75
82
89
96
103 110
Time (days)
Fig.6.2. The residual of sodium oleate in four UASB reactors.
107
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
6.3.2 Identification and quantification of the microbial subpopulations under inhibition
by sodium oleate
In the reactor, acidogens, acetogens and methanogens were expected to mediate conversion
of SO, glucose or cysteine to methane and carbon dioxide. All methanogens are in the
domain Archaea, and acidogens and acetogens are in the domain Bacteria.
The morphological difference of microbial groups before (day19) and after (day33) SO
feeding was visualised on epifluorescence micrographs of FISH using rodamine-dyed
domain probes ARC915 (Fig.6.3). After SO feeding, the long curved rod methanogens
which might be related to the families Methanobacteriaceae and Methanomicrobiaceae
were rare (Fig 6.3a). Moreover, the epifluorescence intensity of Archaea cells before
feeding with SO was stronger than that after feeding with SO, suggesting that the number of
Archaea cells had decreased after feeding with SO. Figure 6.3 (b and c) presents the
morphological differences of order Methanobacteriales (targeted by MB310) and family
Methanosarcinaceae (targeted by MSMX860) after feeding SO. The fluorescent signals or
epifluorescence after feeding SO were all dimmer than before SO feeding with both probes,
and the cell numbers also decreased in all groups of methanogens.
108
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
a
b
c
Fig.6.3 Photographs of different probes with FISH before (day 19, left) and after (day 33,
right) SO feeding. (a) domain Archaea (ARC915). (b) order Methanobacteriales (MB310).
(c) Family Methanosarcinaceae (MSMX860)
Figure 6.4 illustrates the results of the quantification of different subpopulation changes after
feeding SO. Compared to Archaea cell, Bacteria cells suffered less from inhibition by
toxicity of SO. This is in good agreement with previous experiments (Beccari et al., 1996;
109
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
Rinzema et al., 1994). The results also showed that amongst the methanogen groups, family
Methanosarcinaceae cells were the most sensitive to the toxicity of sodium oleate. Because
the family Methanosarcinaceae includes Methanosarcina and Methanosaeta, the only two
genera that have been confirmed so far as acetotrophic methanogens to use acetate, this
might be the main reason why LCFA could severely inhibit the methanogenesis.
12
10
8
6
4
2
After feeding SO
Methanosarcinaceae
Methanomicrobiaceae
Methanocorpusculaceae
Methanoplanaceae
Before feeding SO
Methanobacteriales
Bacteria
0
Archaea
Bcaterial numbers (LogC)
14
Fig. 6.4 The cell numbers in control reactor before (day19) and after (day33) SO feeding.
6.3.3 The re-growth of anaerobic microbial subpopulation during recovery with
cysteine and glucose addition.
The results of the control experiment (Reactor A) indicated that there was no obvious
recovery of both in Archaea and Bacteria groups even after 80 days (Fig. 6.5). The previous
research (Raskin et al., 1994b; Zhang and Noike, 1991) and the data of chapter 4 showed
higher quantity of Archaea cells than Bacteria cells when the anaerobic digester was
operating successfully and was efficiently converting substrates to methane.
In this
experiment, the number of Archaea cells was lower than the number of Bacteria cells even
110
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
after 80 days, there was higher quantity of Archaea cells than Bacteria cells when anaerobic
digester was operating stably and efficiently converting substrates to methane. In this
experiment, the number of Archaea cells was lower than the number of Bacteria cells even
after 80 days, indicating that methanogenic activities were more difficult to recover than
acidogenesis activities from the inhibition of SO (Becarri et al., 1996).
Amongst the
members of Archaea, the re-growth of family Methanosarcinacene was the slowest in these
whole methanogenic groups, suggesting that Methanosarcinaceae cells were more sensitive
to the toxicity of SO.
14
Archaea
Bacteria
12
Methanobacterials
Methanomicrobiaceae Methanocorpusculaceae Methanoplanacea
10
Methanosarcinaceae
8
6
4
2
0
33
43
53
63
73
83
93
103
113
Time (days)
Fig.6.5 . Analysis of microbial populations recovery from LCFA inhibition in control reactor
after 33 days.
In reactor B fed with glucose after SO inhibition, the numbers of Archaea cells immediately
increased during the first 20 days (fig.6.5), and nearly reached the population level before
feeding SO (Fig.6.3). Interestingly, the recovery of number of Bacteria cells was not as
111
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
significant as that of the number of Archaea cells. The members of Methanosarcinaceae had
the best recovery in methanogenic groups after feeding glucose; implying that glucose could
stimulate the acetotrophic methanogens growth effectively, thus accelerate the conversation
of acetate into CH4 and CO2.
In the reactor C fed with cysteine after SO inhibition, the numbers of Archaea did not
increase as fast as when fed with glucose, whereas the numbers of Bacteria increased faster
than when fed with glucose (Fig.6.5). The numbers of Archaea cells as well as the numbers
of methanogens were all lower than that in the reactor which was fed glucose. This could be
linked to the fact that cysteine could not assist as effectively with the recovery of
methanogenesis as glucose (Fig.6.1). Interestingly, the numbers of Bacteria cells were
much higher than those in the glucose fed reactor, in fact even higher than the numbers
before SO inhibition, implying that cysteine benefited the growth of Bacteria cells.
The data of numbers of Archaea and Bacteria cells from feeding of glucose or cysteine,
showed that glucose was more effective to stimulate the growth of Archaea cells than
cysteine, while the cysteine was more beneficial to the recovery of Bacteria cells than
Archaea cells in the reactor under inhibition of LCFA.
112
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
14
a
12
10
8
6
Archaea
4
Bacteria
Methanobacteriales
2
Methanomicrobiaceae Methanocorpusculaceae Methanoplanacea
Methanosarcinaceae
0
33
43
53
63
73
83
93
103
113
Time (days)
14
b
Bacterial numbers (LogC)
12
10
8
Archaea
6
Bacteria
4
Methanobacteriales
2
Methanomicrobiaceae
Methanocorpusculaceae Methanoplanaceae
Methanosarcinaceae
0
33
43
53
63
73
83
93
103
113
Time (days)
Fig.6.6 Recovery of microbial populations under (a) glucose and (b) cysteine feeding.
113
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
Figure 6.7 illustrates that feeding a combination of glucose and cysteine could significantly
increase both the re-growth of Archaea and that of Bacteria cells.
The re-growth of
Bacteria cells was very similar to that observed with feeding cysteine only but the increase
was slower. It might be related to the lower concentration of cysteine in combined feeding
(50% of only cysteine feeding). Although the re-growth of Archaea cells was not as fast as
observed when fed glucose only, the combined feed promoted higher population level than
glucose only after 30 days, suggesting that the combination glucose and cysteine was more
effective in stimulating the growth of Archaea cells than glucose only. The population
levels of re-growth of all methanogenic groups were also higher than those obtained with
from feeding glucose or cysteine only. It could be explained by the fact that both glucose
and amino acid were essential in bacterial growth and they acted as growth factors to
stimulate the growth of microorganisms (Mousa and Forster, 1998).
14
Bacterial numbers (LogC)
12
10
8
Archaea
6
Bacteria
Methanobacteriales
4
Methanomicrobiaceae Methanocorpusculaceae Methanoplanaceae
2
Methanosarcinaceae
0
33
43
53
63
73
83
93
103
113
Time (days)
Fig.6.7 Recovery of different microbial populations fed with glucose and cysteine
114
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
From the experimental data shown in this study, it illustrated that the methanogenic
activities at different feeding were related to the shifts microbial population structure. When
fed with SO, the numbers of Archaea cells and Bacteria cells decreased sharply and this
resulted in the lost of methanogenic activities. With the addition of glucose, the recovery
growth of Archaea cells was higher than that of Bacteria cells and this enhanced the
methanogenic activity. With the addition of cysteine, the recovery growth of Bacteria cells
was higher than Archaea cells and the methanogenic activity was not as high as with glucose
feed, indicating that the high numbers of Archaea cells were more important than those of
Bacteria cells in enhancing the methanogenic activity. When fed with a combination of
glucose and cysteine, high recovery of both Archaea cell and Bacteria cell population were
achieved. Highest methanogenic activity was also achieved with this feeding mode,
suggesting that the combination of glucose and cysteine was most beneficial for the recovery
of methanogenic activities after the inhibition of LCFA.
Although the physico-chemical parameters, such as methane production rate, VFA and pH
were measured to evaluate the effect of organic loading rate and feed composition of the
anaerobic reactor, the microbial parameters could provide more information than physicochemical parameters for understanding disturbance of operation on microbial structure. For
example, when fed combination of glucose and cysteine, the increase of population of
Archaea cells and Bacteria cells in Fig.6.7 could explain the increase of methane production
presented in Fig.6.1. The results of microbial investigation presented detailed reactor
changes in the different feeding modes by microbial population shifts that related to the
efficiency of the reactors and could not be shown by physico-chemical parameters. The
115
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
efficiency of reactor could be predicted and explained by the changes of microbial
communities, and this information could be used to speed up the recovery process by
adjusting the substrate feeding.
Studies presented in Chapters 4, 5, 6 were focused on the effects of glucose and cysteine on
the inhibition of SO using batch or lab-scale reactors. The studies in Chapters 7,8 and 9 will
investigate the function of carbohydrate (starch) and protein (yeast extracts) on the
degradation of lipids (canola oil) using one or two-phase UASB reactors.
These
experiemnts will simulate a real wastewater treatment plant, in which the effluent is likely
to contain mixture of protein, carbohydrates and lipids. The comparison of the efficiency of
the microbial and physico-chemical indicators on the monitoring of the reactor anaerobic
process will also be conducted in Chapter 7.
6.4 Conclusion
Four lab-scale UASB reactors were deliberately inhibited by feeding sodium oleate. Three
of them were fed glucose, cysteine or glucose and cysteine together while one reactor was
operated as a control reactor and was fed continuously with sodium oleate
Two domains
Archaea (targeted by ARC 915) and Bacteria (targeted by EUB 338), one order
Methanobacteriales (targeted by MB310)) and two family- Methanosarcinaceae (targeted
by MSMX860), Methanomicrobiaceae, Methanocorpusculaceae and Methanoplanceae
(targeted by MG 1200) DNA probes were used to quantify the specific groups of anaerobic
microbes throughout the operation of the reactors to compare the effect of glucose and
cysteine on the recovery of reactor. .
116
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
It can be concluded from the methane production rate that glucose had the better effect than
cysteine on recovery the methanogenic activities after inhibition of sodium oleate, while the
combination of glucose and cysteine was most effective on this recovery.
The results of detection and quantification of anaerobic microbial community shifts
provided more detailed information about the recovery process by feeding with glucose,
cysteine or glucose and cysteine. It showed that glucose was more effective in stimulating
the re-growth of methanogen groups than cysteine, while the combination of glucose and
cysteine could achieve the highest recovery of Archaea cells. Feeding cysteine was more
effective than glucose feeding for recovery of Bacteria cells, while the integrated feeding of
glucose and cysteine had highest recovery on Bacteria cells. The numbers of “recovered”
cells of Archaea and Bacteria was nearly the same as those before the SO feeding, implying
that glucose and cysteine had the optimal stimulation on the growth of Archaea and Bacteria
cells. The adaptation or recovery of microbial groups to LCFA might also be related to the
different biological characteristic, nutrition demand of each group during different periods
(Ahring et al., 1992) or other factors still unknown.
From the comparison of physico-chemical parameters and microbial parameters during the
recovery of microbial organisms, it is obvious that microbial study could provide more
detailed information of the recovery process. The knowledge gained from microbial studies
to guide the recovery of different microbial groups from inhibition of lipids or LCFA will
benefit the industry of the anaerobic digestion of lipids containing wastewater, which
117
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 6 Enhancing recovery of methanogenesis from LCFA inhibition by feeding glucose and cysteine in
a single stage UASB reactor
currently encounters frequent shut-down events due to the inhibition of LCFA or lipids.
118
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with
carbohydrate and protein in a one-phase UASB reactor by monitoring the
microbial and physico-chemical indicators
7.1 Introduction
Anaerobic treatment of wastewater containing very high concentration of lipids is a very
difficult task as lipids cause granular sludge flotation and wash-out even at very low organic
loading rates due to the adsorption of lipids by biomass (Rinzema, 1988; Hwu, 1997).
Lipids also cause some engineering problems, such as clogging of gas and effluent lines
(Petruy and Lettinga, 1997). Furthermore, the long-chain fatty acids which are hydrolysed
from lipids can also severely inhibit the methanogenic activity. This was shown in Chapters
4, 5 and 6 and previous reports (Hanaki, 1981; Koster, 1987).
Fongastitkul and Lo (1994) investigated two approaches that could enhance performance of
UASB reactor in treating wastewater with high solid level: (1) increase the density of
bacterial populations in the reactor. (2) increase the bacterial activities by operating at
optimum conditions and supplying essential nutrients for the growth of bacteria. From both
approaches, it could be concluded that the growth of bacteria in the anaerobic reactors was
the key factor to the degradation of complex organic materials in wastewater. To get the
optimal growth of biomass, availability of nutrient is important. Because lipids and LCFAs
119
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
inhibit the growth of biomass, provision of some other organic materials, such as
carbohydrate and protein to support biomass growth during lipid digestion seems to be
reasonable.
Another approach to enhance lipid digestion is to co-digest lipid containing
waste with other organic wastes.
It has been speculated that the co-digestion of different
wastes would supply sufficient nutrients for bacterial growth thus improving the
biodegradability
of
lipid
and
increasing
the
efficiency
of
gas
production
(http://res2.agr.ca/initiatives/manurenet/en/man).
Although some work has been performed in lipid anaerobic digestion, its evaluation was
limited to the physico-chemical parameters, like on-line or off-line investigation of gas
composition, pH, COD and VFA (Rinzema, 1988; Hwu, 1997; Beccari et al., 1996) and the
microbial analysis was usually not performed due to the difficulties to apply microbiological
techniques. The physico-chemical parameters were used to control digester or diagnose
problems in the digestion process.
Some research has been carried out on investigating
biomass in anaerobic process but indirectly through monitoring the metabolic products
related to specific bacterial groups. However, the results were not always reliable because
different bacterial groups could produce the same products through different process and at
different operating conditions (Rozzi and Remigi, 2002).
Raskin et al. (1994a) reported that knowledge of microbial composition is necessary and
may be used as a direct indicator of performance of anaerobic process.
But in most
wastewater plants, the routine monitoring of anaerobic digestion is performed by analysing
120
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
physicochemical indicators, such as total solid, VFA, alkalinity, pH and composition of
biogas. In theory, the effect of a disturbance in the process on the microbial population will
be seen before the changes of physico-chemical indicators. The changes of these phsicochemical indicators are slower than biological indicators and may not be effective for reactor
monitoring (Farrell et al., 1988). Primary disturbances that cause failure of anaerobic
digestion process are feed overload and presence of inhibitors in feed (Rinzema, 1988). The
disturbances may cause the microbial populations to decrease severely that they may have
difficulty to recover. If only physico-chemical indicators are applied for monitoring, it will
sometimes be very hard to control the failure of the reactor. Thus monitoring the changes in
microbial community composition may be beneficial.
However, the application of biomass
monitoring seems impractical due to the difficulties associated with traditional microbial
investigation.
The hybrid UASB reactor of Spearwood Wastewater Treatment Plant (WA, Australia)
treating slaughterhouse wastewater was shut down in 1997 due to overloading of wastewater
from slaughterhouse. A long period was required to re-start the reactor. The method used
for analysing bacterial activity was to measure the gas production after shaking the sludge
sample in a closed plastic container for several minutes. The growth and composition of the
microbial community was treated as a black box. Fortunately, with the recently developed
molecular techniques, which were discussed in the previous chapters, the application of
small-subunit (SSU) rRNA probes to investigate the sophisticated microbial communities in
the reactor is possible. Application of these techniques will provide an in-depth indication of
121
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
the composition of microbial community. Because all methanogen groups presented in
anaerobic digestion are included in the domain Archaea, while acetogens and acidogens are
included in domain Bacteria, domain oligonucleotide hybridisation probes were used in the
experiment to monitor these groups.
The objective of this chapter was to evaluate the effect of co-digestion of protein and
carbohydrate on lipid treatment in the one-phase UASB anaerobic reactor by monitoring the
physico-chemical indicators and the shifts of bacterial population. From the results obtained
a comparison of the effectiveness of monitoring biological parameters over physicochemical indicators will be possible.
7.2 Materials and methods
7.2.1
Feeding conditions and sampling
Two, one-phase UASB reactors were set up as described 3.1.2 and were seeded with
granular sludge from the hybrid UASB reactor of Spearwood Wastewater Treatment Plant
(Australia,WA). The hybrid UASB reactor treats wastewater from a pig abattoir and dairy
processing plants which are high in fat and protein. Therefore, it was expected that the
granular sludge will easily acclimatize to the lipid (canola oil), protein (yeast extracts) and
carbohydrate (starch) feed in the studies.
122
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
One of UASB reactors was only fed with canola oil and set up as a control reactor. It was
operated in parallel to another UASB reactor. This other UASB reactor was fed with
varying composition of canola oil, starch and yeast extract in different runs (as shown in
Table 7.1). Before feeding, both reactors were reactivated by 2 gCOD L-1 day
-1
of starch
for 1 week to obtain stable methanogenic activity. Each run lasted 40 days before shifting to
the next run.
Sampling of sludge for physico-chemical and microbial assays in each run
was taken every day for the first 5 days, and then taken once every 5 days. The reactor was
operated at a hydraulic retention time (HRT) of 4 days and the loading rate was kept at 2
gCOD L-1 day -1. The temperature of the reactor was maintained at 35 oC using a water
jacket connected to a water bath.
Component
Control
Run1
Run2
Run3
Run4
Run5
Run6
Starch
0
60
50
40
30
20
10
Yeast extract 0
30
30
40
40
40
40
Canola oil
10
20
20
30
40
50
100
Table 7.1 Feed composition for the different runs expressed as the % of 2 gCOD. L-1 day –1.
Note: 1 g carbohydrates (starch) is equivalent to 1.01 g COD; 1 g lipid (canola oil) is equivalent to 2.86 g
COD; 1 g protein (yeast extract) is equivalent to 1.35 g COD
123
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
7.2.2
Monitoring of Physico-chemical indicators
The physico-chemical parameters monitored were total VFAs (acetate, propionate, butyrate)
concentration and daily methane production. They were analysed by the methods described
in 3.6.
7.2.3
Microbial analysis of UASB reactor
The dynamic changes of Eubacteria and Archaea during the six runs were analysed by dual
staining the sludge samples by FISH and DAPI methods.
The FISH method which was as described in 3.4 and the DAPI (Hicks et al., 1992) method
were applied to identify and quantify the population changes during the different runs.
7.2.4
Analysis of lipid, carbohydrate and protein
Methods for analysis of oil, starch and yeast extract were as described in the section 3.7.
The extent of the adsorption of oil on sludge was determined separately by setting up batch
assays containing the same amount of sludge as in the UASB reactor, except that the sludge
was autoclaved. The adsorptions of yeast extract and starch were determined with the same
method. The extent of the degradation of each component (g COD/L) was calculated
according to:
124
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
Total component – Residual component – Adsorbed component
Total component
7.3
Results and discussions
7.3.1 Lipid inhibition on UASB reactor
In the control experiment, the addition of canola oil (2 gCOD. L-1 day
–1
) caused severe
inhibition of methanogenic activity. As shown in Fig.7.1 the methane production rate
dropped to zero within 3 days of the oil being introduced. This inhibition lasted for the
duration of experiment. Lipid analysis showed that there was about 10% degradation of
lipid during the first 5 days, and less than 5 % thereafter, suggesting that canola oil did not
only inhibit the methanogenic activities, but also inhibited the acidogenic activities as
reported previously (Hanaki et al., 1987). The numbers of Archaea and Bacteria cells were
very low during the whole process, around 10 3 and 10 6 /ml, respectively. As reported by
Hwu (1997) and Rinzema (1988), the oil clogged the gas and effluent lines of the reactor
and the reactor contents turned to a dark brown colour (Hwu, 1998; Rinzema, 1988).
125
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
Methane production
(ml/l.day)
600
500
400
methane
300
200
100
0
0
50
100
150
200
250
Time (days)
Fig. 7.1 Methane production rate in the control reactor.
7.3.2 The influence of feed compositions on digester performance.
Figures 7.2 and 7.3 show the methane production rate and the total VFA during the
operation of UASB using various feed compositions. Upon introducing lipid into the reactor
along with starch and yeast extract after the initial reactivation with starch alone, methane
production and VFA both decreased even though the total loading rate was kept the same.
This indicated that the higher loading rate of oil or yeast extract decreased the methanogenic
and acidogenic activities. When reactor was shifted from 100% starch feeding to starch,
yeast extract and oil (Run1, ratio 6:3:1 in w:w:w), the methane production and extent of
degradation of starch, yeast extract and oil all slightly decreased (Fig.7.2). When oil feeding
was increased from 10 to 20%, starch feeding decreased from 60 to 50% and yeast extract
was kept same (Run1 to Run2), the methane production decreased from 500 to 325 ml/l.day,
while the production of VFA decreased from 253 to 181 mg/l and extent of degradation of
126
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
oil decreased 45 to 26%. This suggested that the increase of oil concentration decreased both
the acidogenic and methanogenic activities of the UASB reactor. From Run 2 onwards, the
methane production continuously decreased, this could be attributed to the increase of the
loading rate of yeast extract or oil and the decrease of starch. However, it is more likely that
20% loading rate of oil exceeded the threshold oil loading rate for the UASB reactor and
from 40 days i.e. beginning of Run2, the biomass was being washed out from the digester
and leading to a decrease of methane production. The extents of degradation of starch and
yeast extract also continuously decreased, suggesting that higher concentration of oil
inhibited the degradation starch or yeast extract.
Interestingly, with the decrease of methane, VFA concentration decreased continuously
instead of increasing. This could be attributed to the inhibition of oil on acidogenesis and
acetogenesis as well (Fig.7.3). The decrease of acidogenesis and acetogenesis caused a
decrease of VFA production.
The LCFA concentrations at Run1 were the lowest. It
increased with the increase of oil loading rate indicating that lower oil loading rate in codigestion benefited the LCFA degradation or transformation. Because high concentration of
LCFA is toxic to the methanogens as well as to the acidogens, lower methane production
and VFA were produced in the high loading rate runs.
Because starch (60%), yeast extract (30%) and oil (10%) were in total 2g COD/l.day
feeding, it could be calculated that 1.2, 0.6 and 0.2 g COD /l.day starch, yeast extract and oil
were fed to UASB reactor, respectively. From the results of the extent of degradation of
127
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a
one-phase UASB reactor by monitoring the microbial and physico-chemical indicators
starch (85%), yeast extract (73%) and oil (45%) at Run1 (mean value), it could be calculated
that 1.02, 0.43 and 0.09 g COD/l.day hydrolysed from starch, yeast extract and oil,
respectively. If all the hydrolysed products proceeded to methane, about 557 ml methane
could be produced per day. This value is very near that of methane produced in the Run1
(502 ml/day), suggesting that only acidogenesis was depress in Run1. The methanogenesis
rate could be calculated as the methane production (502ml/day) in the reactor divided by
theoretical methane production (557ml/day), i.e., 90.1%. These calculations were performed
for all runs and shown at table 7.2. It showed that with the increase of oil and yeast extract
loading rate, the methanogenesis rate decreased. Taking the LCFA results (Fig.7.5) in
conjunction with the methane production data suggested that a higher loading rate of
carbohydrate in co-digestion would benefit the methanogenic and acidogenic or acetogenic
activities.
Run1
Run2
Run3
Run4
Run5
Run6
Theoretical methane production (ml/day)
557
407
283
217
140
70
Methane production (ml/day) in UASB
502
353
232
135
46
0
Methanogenesis rate (%)
90.1
86.7
81.9
62.2
32.8
0
Table 7.2. Extent of methanogenesis for different runs
128
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a one-phase UASB reactor by monitoring the microbial and
physico-chemical indicators
600
Methane production (ml/day)
550
500
450
Run 1
400
Run 2
Run 3
Run 4
Run 6
Run 5
350
300
250
200
150
100
50
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Time (days)
Fig. 7.2 Methane production rate for the different runs.
129
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a one-phase UASB reactor by monitoring the microbial and
physico-chemical indicators
300
VFA
Run1
250
Run2
Run3
Run4
Run5
Run6
200
150
100
50
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Days
Fig.7.3. VFA concentrations for the different runs.
130
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a one-phase UASB reactor by monitoring the microbial and
physico-chemical indicators
Percentage of degraded component
100
Run 1
90
Run 2
Run 3
Run 4
Run 5
Run 6
Canola oil
80
starch
70
Yeast extract
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Days
Fig.7.4. Degradation of three organic components in feed for the different runs
131
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a one-phase UASB reactor by monitoring the microbial and
physico-chemical indicators
LCFA concentration (mg/l)
10
Run 1
9
Run 2
Run 3
Run 5
Run 4
Run6
8
7
6
5
4
3
2
1
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Time (days)
Fig.7.5 LCFA hydrolysed in the UASB rector for the different runs.
132
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a one-phase UASB reactor by monitoring the microbial and
physico-chemical indicators
16
Cell numbers (LogC)/ml
14
12
10
Run 2
Run 1
8
Run 3
Run 4
Run 5
Run 6
6
4
Archaea
2
Eubacteria
0
0
10
20
30
40
50
60
70
80
90
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240
Time (Days)
Fig.7.6 Comparison of the shifts of Bacteria and Archaea population for the different runs.
133
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a onephase UASB reactor by monitoring the microbial and physico-chemical indicators
7.3.3 Microbial communities shifts under different feeding codes
Figure 7.6 shows that the numbers of Archaea cells were larger than the numbers of
Eubacteria cells when only 10% of oil was in feed (Run 1). However, when increasing the
oil ratio from 10% to 20% from Run 1 to Run 2, the numbers of Archaea cell decreased
sharply. However, it was kept stable from Run 2 to Run 3 when the oil loading rate was the
same even though the loading rate of starch decreased and that of yeast extract increased.
These suggested that the numbers of Archaea cells were only related to the concentration of
oil, and that the concentrations of starch and yeast extract were less important. With the
increase of oil loading rate from Run 4 to Run 6, the numbers of Archaea cells decreased,
indicating that higher oil loading rate or oil concentration did not benefit the growth of
Archaea. This is in accordance with the methanogenesis rate shown in the Table 7.2 in
which the methanogenesis rate decreased significantly from Run 4 to Run 6.
Unlike Archaea cells, Bacteria cell were less affected by oil addition. This is in good
agreement with previous studies (Galbraith and Miller, 1973) and experiment results
described in the Chapter 4. Archaea member were more sensitive to the inhibition of LCFA
than Bacteria. Although both Bacteria and Archaea cell numbers decreased when the oil
concentration in feed was increased from Run 3 to Run 4, the numbers of Bacteria cells
recovered a significantly in both runs while the numbers of Archaea cells were inhibited
severely by oil, indicating that Bacteria cell was less sensitive to the toxicity of oil than
Archaea cell. However, even Bacteria cells demonstrated some recovery from Run 5 to
134
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a onephase UASB reactor by monitoring the microbial and physico-chemical indicators
Run6, their numbers could not reach the level of Run1, suggesting that too high oil loading
would also inhibited the growth of Bacteria cells and the recovery of Bacteria cell would
not be completed. There was not only significant recovery of Archaea cell in the all runs,
suggesting that the recovery of Archaea cells was a long process and more difficult than the
recovery of Bacteria cell.
7.3.4 Comparison of the efficiency of physico-chemical and microbial monitoring
The data depicted in Fig. 7.2 (methane production rate), Fig.7.3 (VFA concentrations),
Fig.7.4 (extent of degradation of organics) and Fig.7.5 (LCFA concentrations) showed that
physico-chemical parameters responded slowly when the loading rate of oil was changed.
However, the shifts of bacterial population, especially for Archaea cells were significantly
faster in the same conditions. After feeding 10% oil (Run1), the Archaea cell numbers
exhibited a Log 2 reduction after 18 hours only, while the methane production rate only
decreased by 50ml /day after 5 days. The VFA concentrations and the extent of organics
degradation did not change appreciably. A Log 4 reduction of the Archaea cell numbers
was observed when the oil loading rate increased from 20% to 30% (from Run3 to Run 40)
compared to a decrease of less than 10 mg/l in the VFA concentration in 5 days.
This
indicated that investigating Archaea cell population shifts would be more effective than
were observing the changes of physico-chemical indicators when toxic organic wastes were
fed. By the time the significant changes in physico-chemical indicators were observed, the
135
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a onephase UASB reactor by monitoring the microbial and physico-chemical indicators
anaerobic digester would have been severely inhibited and difficult to recover. However,
there are still some difficulties in monitoring biological parameters. These techniques are
time consuming and require a skilled microbiologist to perform assay whereas the plant
operator can be trained to carry out routine analysis of physico-chemical parameters. Apart
from this, it seems more difficult to apply biological indicator monitoring techniques on-line
or off-line than physico-chemical parameters due to the complexities in bacteria
identification and quantification.
7.3.5 Efficiency of one-phase UASB in the degradation of oil combination with starch
and yeast extract feeding.
The addition of starch and yeast extract significantly enhanced oil methanogenesis in onephase UASB reactor when the oil loading rate was lower than 30% in a total of 2
gCOD/l.day. However, less than 50% of the oil was decreased when co-digested with starch
and yeast extract even at the lower concentration (10%), and that percentage decreased with
the increase of oil concentration. The VFA concentration decreased continuously.
It
indicated that one-phase UASB reactor had relatively low capacity to degrade oil even
though co-digested with starch and yeast extract.
Furthermore, the increase of oil
concentration decreased the degradation of both starch and yeast extract. Among the three
organic components, starch had the best degradation in all runs.
7.4 Summary and conclusion
136
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 7 Evaluation of lipid anaerobic digestion when co-digested with carbohydrate and protein in a onephase UASB reactor by monitoring the microbial and physico-chemical indicators
The physicochemical and biological monitoring of oil degradation combined with starch and
yeast extract in one-phase UASB digester was investigated in this experiment. The results
showed that the monitoring of biological parameter, such as bacterial numbers to monitor
the oil degradation process with different feeding modes was as effective as that of
physicochemical parameters, such as VFA, methane production rate and extent of
degradation of organic components. Biological monitoring could avoid digester failure
resulting from wrong operation by providing early warning. Further research to develop the
simple microbial methods for on-line biological parameters monitoring is necessary.
The results of this experiment also demonstrated that the application of one-phase UASB
reactor was not beneficial for oil degradation either when oil was fed by itself or in
combination with yeast extract and starch. This application of two-phase UASB reactor to
enhance degradation of oil when co-digested with protein and carbohydrate will therefore be
investigated in the Chapters 8 and 9.
137
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase
anaerobic treatment systems.
8.1 Introduction
Previous investigations have demonstrated that the hydrolysis of lipid to LCFA and glycerol
was a kinetic bottleneck of the anaerobic digestion process (Beccari et al., 1996; Petruy and
Lettinga, 1997). Lipid hydrolysis could be enhanced by adding emulsifying chemicals, such
as the anionic surfactant sodium dodecylbenzenesulfonate (SDS) before hydrolysis (Charles
et al, 1996). This treatment reduces the size of the lipid particles, providing more reactive
surface area for hydrolysis (Petruy and Lettinga, 1997). However, these chemicals have
been shown to be toxic to anaerobic bacteria (Petruy and Lettinga, 1997).
Yet another approach to enhance lipid hydrolysis is the use of two-phase anaerobic digestion
systems which are becoming increasingly popular for wastewater treatment. These consist
of two serially connected reactors, with the first reactor known as an acidogenic reactor and
the second the methanogenic reactor.
In the first reactor the organic matter in the
wastewater is hydrolysed or fermented to organic acids. This reactor is normally operated at
short hydraulic retention times, which ensures that little or no methanogenesis occurs. Since
Ghosh and Pohland (1974) first proposed two-phase reactor system, many efforts had been
focused on the studies of physico-chemical parameters in acidogenic and methanogenic
reactors during anaerobic digestion of the complex organic materials (Yilmazer and
Yenigun, 1999; Jeyaseenlan and Matsuo, 1995). However, the shifts of bacteria populations
138
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
in the two reactors under different experimental conditions is less known. It has been
demonstrated that two-phase reactors for lipid anaerobic digestion had the apparent
advantage over one-phase reactor in that separation of phases could decrease the inhibition
of LCFA on methanogenesis (Hanaki et al., 1987; Becarri and Torrisi, 1998). Effluent from
the acidogenic reactor was then pumped into the second reactor where methanogenesis
mainly occurred. Inhibition of lipid on methanogenic activity could be attributed to the
unsaturated LCFAs. Saturated LCFAs were shown not to cause any inhibition on
methanogenesis (Komatsu et al., 1991). The acidogenic reactor in two-phase reactor system
could saturate unsaturated LCFAs due to the presence of a modest hydrogenotrophic
activities (Becarri and Torrisi, 1998). Two-phase anaerobic digesters have been used
successfully to treat effluent from olive oil mill and cheese whey (Beccari et al., 1996;
Petruy nd Lettinga, 1997; Yilmazer and Yenigun, 1999) containing up to 200 g COD/l in
lipids. The extent of lipid digestion in a two-phase operation was shown to be twice as that
in a single-phase operation (Jeyaseelan and Matsuo, 1995). Hanaki et al. (1987) found that
even though the performance of two-phase and single-phase system were similar when
treating carbohydrates and proteins, the two-phase systems was more effective in preventing
inhibition by LCFAs when treating lipid. It was suggested that LCFAs were transformed
into less toxic forms (i.e. transformation of unsaturated LCFA to saturated LCFA) by
acidogenic bacteria in the first reactor thereby preventing inhibition in the second
(methanogenic) reactor (Komatsu et al., 1991). Compared to mesophilic temperatures,
operation at thermophilic temperatures (> 55oC) has also been shown to improve breakdown
of lipids in the acidogenic reactor (Ibrahim et al., 1984).
139
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
The study presented here evaluates the effect of a combination of parameters, like
temperature, HRT and feed type, on the extent of hydrolysis of lipids in two-phase anaerobic
treatment system. Following the lipid degradation studies in one phase anaerobic reactor
discussed in Chapter 7, the effects of addition of protein and carbohydrate
on both
acidogenic and methanogenic reactors operated at various hydraulic retention times and
mesophilic (35oC) or thermophilic conditions (55oC) on the extent of hydrolysis of lipid
were investigated. Canola oil was used as the model lipid and yeast extract and starch as
protein and carbohydrate respectively. The Archaeal and Bacterial communities in the
acidogenic reactor were enumerated using rRNA fluorescence in-situ hybridization (FISH)
molecular technique.
8.2 Materials and methods
8.2.1. Acidogenic and methanogenic reactors and their operation. Two sets of lab-scale,
two-phase UASB processes were set up in parallel. Each two-phase UASB process was
made up of two Perspex, water-jacketed bottles each of 2 litre working volume and served
as the acidogenic and methanogenic reactor. Both acidogenic and methanogenic reactors of
one set was operated at 35oC, while the other was maintained at 55oC. The acidogenic
reactor was inoculated with 1 litre of fine sludge from the acidogenic stage (equalisation
tank) and the methanogenic reactor was seeded with 1 litre granular sludge from a full-scale
hybrid mesophilic UASB reactor at the Spearwood Wastewater Treatment Plant (Australia,
WA). The inoculum was stored at 4oC prior to use.
140
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
A set of experiments was performed in parallel on mesophilic and thermophilic system.
Each set included four runs (Table 8.1). Each run lasted 3 months.
As listed the
concentrations of yeast extract, starch and canola oil in the feed for the different runs were
adjusted in such a way that the total COD of the feed was 2 g COD.L-1. Canola oil was
emulsified by sonication with an ultrasonic clearer (Branson, Sonifier 450) before feeding
and was kept at 1g COD in run1. The feed was refrigerated at 5 0 C and fed to the reactor by
a peristaltic pump. During each experimental run the reactor was operated at 1, 2, 3, 4 and
5 day hydraulic retention time. The reactor was maintained at each hydraulic retention time
for two to three weeks. The inoculum in the reactor was changed to the original inoculum
from the full-scale treatment plant at the end of run so as to keep the starting inoculum for
all runs the same.
Run1
(mgCOD/l
canola oil
Run2
(mgCOD/l)
Run3
Run4
(mgCOD/l)
(mgCOD/l)
1000
1000
1000
1000
starch
0
1000
0
500
yeast extrac
0
0
1000
500
Table 8.1. Feed composition for different runs.
8.2.2 Enumeration of Archaeal and Bacterial communities. The microbial population
shifts of each run under different HRT and temperature were characterised by FISH
141
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
techniques. Analyses were conducted in triplicate and data were averaged. The samples
from the acidogenic reactor were taken at the same level of the reactor.
Two
oligonucleotide probes, ARC915 (Stahl et al., 1991) and EUB338 (Amann et al., 1990a)
which targeted to Archaea and Bacteria cells were applied to identify and quantify these
organisms.
8.2.3 Measurement of extent of hydrolysis of lipid, protein and carbohydrate. The
analysis for the quantification of starch and yeast extract and canola oil was performed
according to the methods described in sections 3.7.1, 3.7.2 and 3.7.3. The concentration of
canola oil was determined gravimetrically after extraction of lipid from samples by
petroleum ether, according to the Soxlett extraction method (APHA, 1985). The anthrone
method was applied for the analysis of starch and the Lowry method for the analysis protein
(Raunkjaer et al., 1994). The extent of the adsorption of lipid on sludge was determined by
subtracting the liquid phase lipid from the total lipid that was added to 1 litre of autoclaved
sludge from the full-scale plant.
The extent of hydrolysis (%) of canola oil, starch and protein for each run were calculated
using:
Hydrolysis (%) =
Total amount added − Re sidual amount − Adsorbed amount
Total amount added
The organic materials present in sludge before addition of substrates were assumed to be
zero.
142
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
8.2.4 Analysis of LCFA hydrolysed from canola oil. The four main LCFA, oleic acid
(C18:1), linoleic acid (C18:2), linolenic acid (C18:3), and palmitic acid (C16:0) in
canola oil were analysed by the method described in 3.7.1.
The transformation or
saturation of individual LCFA in acidogenic reactor was evaluated by comparing their
concentrations in different experimental conditions.
8.3 Results and discussion
It should be noted that there was no methane gas produced in the acidogenic reactor in the
studies reported here.
8.3.1 Effects of HRT on hydrolysis of lipid, carbohydrates and protein in acidogenic
reactor.
At mesophilic temperature (Fig.8.1a), the extent of hydrolysis in acidogenic
reactor which was only fed with canola oil was between 6-9% and it appeared to be
independent of the HRTs investigated. The highest extent of hydrolysis of canola oil was
measured when oil was fed along with starch and varied between 26 and 35%.
143
Lipid hydrolysis (percentage)
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
70
oil
60
oil+starch
50
a
oil+YE
40
oil+YE+starch
30
20
10
0
1
2
3
4
5
Lipid hydrolysis (percentage)
HRT (days)
70
oil
60
oil+starch
50
b
oil+YE
40
oil+YE+starch
30
20
10
0
1
2
3
4
5
HRT (days)
Figure 8.1. Hydrolysis of canola oil in acidogenic reactor at different feed composition and
HRT at (a) mesophilic and (b) thermophilic condition.
The extent of hydrolysis of canola oil varied between 10 and 14 % when fed with yeast
extract, and 22 and 27% when fed with starch and yeast extract, suggesting that starch had
the better effect on enhancing the hydrolysis of canola oil. Interestingly, increasing HRT
beyond 1 day did not have an apparent effect on the extent of hydrolysis when oil was fed
144
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
with co-substrates, implying that one day HRT was sufficient for acquiring the maximum
extent of hydrolysis of canola oil in co-digestion with yeast extract and starch.
At thermophilic temperature (Fig. 8.1b), the extent of hydrolysis of canola oil was greater
than that at mesophilic condition for all feed combinations and HRTs. The extent of
hydrolysis of oil without the presence of any other co-substrates feeding was between 5 and
7%. However, when fed with starch, it increased to 55-63%, with yeast extract it increased
to 33-38%, and in combination with yeast extract and starch it was around 48%. The higher
hydrolysis achieved in thermophilic condition could be attributed to two facts. One is that
the high temperature benefited the growth rate of the microorganisms, especially
thermophiles growth (Hwu, 1997). Even though the numbers of thermophiles in the original
mesophilic inoculum were very low, they may have become very active at high temperature
and thus degraded the lipid effectively. The other reason might be related to the fact that
biosorption of lipid increased with the increase of temperature (Riffat and Dague, 1995).
They reported that with the increasing of biosoption, the COD removal in the effluent was
also increased. The extent of biosorption will be discussed later.
The actual reason for the advantage of co-digestion in lipid degradation is still unclear and it
could be related to the increase in the lipid surface area due to the presence of protein and
carbohydrate and their fermentation products might better emulsify the lipids. It is known
that an increase of particulate surface will enhance the hydrolysis efficiency (Zeeman and
Sanders, 2001). Or it might be related to the fact that the complex organics could provide
145
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
the more complete essential nutrients for the growth of the hydrolysis acidogens which will
be discussed in the microbial experiment of this chapter.
In mesophilic condition, starch (2g COD/l) fed by itself was mostly hydrolysed. The
hydrolysis extent was 98 % at HRT of 1 day and 99 % at HRT of 5 days. High hydrolysis
of starch could be achieved at short HRT (< 1 day). This is in good agreement with previous
reports (Hanaki et al., 1987; Pavlostathis and Giraldogomez., 1991). The easy hydrolysis of
starch might be explained by its high solubility. However, the extent of hydrolysis of starch
decreased when it was fed with oil or oil and yeast extract at all HRTs (Fig.8.2). At an HRT
of 1 day, the lowest percentage of starch hydrolysis measured was 85% and 75% by feeding
with oil and oil together with yeast extract, respectively. The hydrolysis of starch increased
when HRT increased from 1 day to 2days both when only oil or oil together with yeast
extract was fed. When HRT increased from 2 days to 5 days, the extent of starch hydrolysis
varied between 95 to 91 % by feeding with oil, while it increased from 87 to 94 % when oil
together with yeast extract was added.
Starch (2g COD/l) was mostly hydrolysed at thermophilic temperature when fed alone and
was irrespective of HRT. However, the extent of hydrolysis of starch was lowered and
slightly decreased with increase the HRT when fed with oil or oil and yeast extract. At an
HRT of 1 day, the hydrolysis extent was 86% when fed with oil and 83% when fed with oil
and yeast extract. These figures decreased to 70% and 71% respectively at HRT of 5 days.
The presence of oil clearly inhibited the extend of starch hydrolysis even though the addition
146
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
of starch increased the extent of hydrolysis of oil. The effect on microbial population will be
Starch hydrolysis (%)
discussed later.
Starch
100
a
Starch+oil
80
Starch+oi+YE
60
40
20
0
1
2
3
4
5
Starch hydrolysis (%)
HRT (days)
Starch
100
b
Starch+oil
80
Starch+oi+YE
60
40
20
0
1
2
3
4
5
HRT (days)
Fig.8.2 Hydrolysis of starch in acidogenic reactor at different feed compositions and HRT at
(a) mesophilic and (b) thermophilic temperature.
In mesophilic condition, the extent of hydrolysis of protein in yeast extract (2g COD/l) when
fed without starch or oil was between 76 and 80% and was independent of HRT when HRT
≥ 2 days. However, the extent of hydrolysis of protein was greatly reduced when fed with oil
147
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
(56 –58%) and with oil and starch (65-69%) (Fig.8.3).
The hydrolysis extent of protein
when fed with oil and starch was higher than when just fed with oil, suggesting that starch
Yeast extract hydrolysis
(%)
could decrease the inhibition of oil on the hydrolysis of protein.
YE
100
a
YE+oil
80
YE+oil+starch
60
40
20
0
1
2
3
4
5
Yeast extract hydrolysis
(%)
HRT (days)
YE
YE+oil
YE+oil+starch
100
80
b
60
40
20
0
1
2
3
4
5
HRT (days)
Fig. 8.3 Hydrolysis of yeast extract in acidogenic reactor of different fed compositions and
HRT at (a) mesophilic and (b) thermophilic temperature.
The extent of protein hydrolysis in yeast extract (2g COD/l) was greater (> 80%) at
thermophilic temperature than at mesophilic condition. It slightly increased from 83% to 87
% when HRT was increased from 1 day to 5 days. Interestingly, at thermophilic
148
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
temperature, the hydrolysis of protein in yeast extract with addition of oil (71-75%) was
higher than that when both oil and starch was fed (65-68%). The results might be attributed
to the fact that different temperature and nutrient composition could benefit the growth of
specific bacteria groups, which in return, benefited the hydrolysis.
Fig.8.2 and 8.3 show that the addition of oil inhibited hydrolysis of both yeast extract and
starch. It might be related to the fact that lipids and their hydrolysed products LCFA were
toxic to the growth of microoganisms.
8.3.2 Formation and transformation of LCFA in acidogenic reactor.
LCFA and
glycerol are expected products in the first step of canola oil hydrolysis. These are further
transformed into other short chain fatty acids or VFA.
The LCFA, oleic acid (C18:1),
linoleic acid (C18:2), linolenic acid (C18:3) palmitic acid (C16:0) were the primary LCFA
of the canola oil used in the experiment. Upon complete hydrolysis of 1 gCOD/l of canola
oil, 320mg/l of LCFA are expected to be produced. In the acidogenic reactor, the expected
LCFA concentration could be calculated by multiplying the extent of oil hydrolysed with the
amount of LCFA in oil. The extent of LCFA transformation was estimated by subtracting
measured LCFA from the expected amount of LCFA then dividing the difference by the
expected amount of LCFA.
Figure 8.4 shows that the addition of starch and yeast extract stimulated the transformation
of LCFA which was hydrolysed from oil in acidogenic reactor. At mesophilic temperature,
the extent of LCFA transformation in acidogenic reactor when fed with starch, yeast extract
149
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
or starch and yeast extract were very similar (18 to 20%) at different HRT and slightly
increased upon increasing HRT from 1 to 2 to 3 days. At thermophilic temperature, the
extents of LCFA transformation in different feeding modes were very similar at different
HRT, indicating that the LCFA transformation was independent of HRT at thermophilic
temperature. The results showed that the transformations of LCFA in acidogenic reactor at
mesophilic and thermophilic temperature were very low, suggesting that the LCFA
transformation in acidogenic reactor was suppressed during the oil degradation in two-phase
anaerobic process.
oil
Extent of LCFA
transformation (%)
30
a
oil+starch
25
oil+yeast
20
oil+yeast+starch
15
10
5
0
1
2
3
4
5
HRT (days)
oil
Extent of LCFA
transformation (%)
25
b
oil+starch
20
oil+yeast
15
oil+yeast+starch
10
5
0
1
2
3
4
5
HRT (days)
Fig. 8.4. LCFA transformation at (a) mesophilic and (b) thermophilic temperature.
150
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
The residual concentration of the four LCFAs: oleic acid, linoleic acid, linolenic acid and
palmitic acid in acidogenic reactor was higher at thermophilic temperature than at
mesophilic temperature (Fig.8.5). The concentration of residual LCFA was the highest when
the reactor was fed by oil and starch, while the concentration of LCFA was the lowest when
it was only fed by oil both at mesophilic and thermophilic temperature. This was in
accordance with the results shown in Fig.8.1 in which the oil was best hydrolysed when oil
and starch was fed in acidogenic reactor, while it was hardly hydrolysed when only oil was
fed.
In the acidogenic reactor, the concentrations of the four LCFA were detected at different
HRT, feed compositions and temperature. During the hydrolysis, unsaturated LCFA, such as
oleic acid (C18:1), linoleic acid (C18:2), linolenic acid (C18:3) might be saturated into
saturated LCFA, such as, stearic acid (C18:0), palmitic acid (C16:0) or other saturated
LCFA. Because oleic acid (61%), linoleic acid (21%), linolenic acid (9%) and palmitic acid
(4%) made up the LCFA in canola oil used in the experiment, after hydrolysis, the ratio
among oleic, linoleic, linolenic and palmitic acids were expected to be at 61:21:9:4 ratio,
assuming no further transformation of LCFA occurs or the transformation of each of LCFAs
was in the same proportion.
151
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
Residual LCFA w hen oil w as fed
Residual LCFA w hen oil w as fed
Linoleic acid
40
Oleic acid
20
Palmitic acid
0
1
2
3
4
LCFA
concentration
(mg/l)
LCFA
concentration
(mg/l)
Linolenic acid
60
Linolenic acid
60
Linoleic acid
40
Oleic acid
20
Palmitic acid
0
5
1
Residual LCFA w hen oil and starch
w ere fed
Linolenic acid
200
Linoleic acid
150
Oleic acid
100
50
Palmitic acid
0
1
2
3
4
Oleic acid
100
50
Palmitic acid
0
1
Linoleic acid
80
Oleic acid
40
Palmitic acid
0
Linolenic acid
Linoleic acid
Oleic acid
40
Palmitic acid
0
1
Linoleic acid
80
40
Oleic acid
Palmitic acid
0
HRT (days)
3
4
5
5
Residual LCFA w hen oil, starch and
yeast extract w ere fed
LCFA
concentration
(mg/l)
LCFA
concentration
(mg/l)
Linolenic acid
120
4
2
HRT (days)
160
3
5
80
5
Residual LCFA w hen oil,starch and
yeast extract w ere fed
2
4
120
HRT (days)
1
3
Residual LCFA w hen oil and yeast
extract w ere fed
LCFA
concentration
(mg/l)
LCFA
concentration
(mg/l)
Linolenic acid
4
2
HRT (days)
120
3
5
150
5
Residual LCFA w hen oil and yeast
extract w ere fed
2
4
Residual LCFA w hen oil and starch
w ere fed
Linolenic acid
200
Linoleic acid
HRT (days)
1
3
HRT (days)
LCFA
concentration
(mg/l)
LCFA
concentration
(mg/l)
HRT (days)
2
160
Linolenic acid
120
Linoleic acid
80
40
Oleic acid
Palmitic acid
0
1
2
3
4
5
HRT (days)
Fig.8.5. Concentrations of individual LCFA under different feed compositions at mesophilic
(left column) and thermophilic temperature (right column).
152
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
Fig.8.5 indicated that the ratio of palmatic acid to the four LCFA increased from 4% to 36%
(oil and starch feeding), 38% (oil and yeast extract feeding) and 25% (oil, starch and yeast
extract feeding) at mesophilic temperature, and 67% (oil and starch feeding), 24% (oil and
yeast extract feeding) and 29% (oil, starch and yeast extract feeding) at thermophilic
temperature. There was a significant saturation of unsaturated LCFA to palmitic acid in
acidogenic reactor when fed oil with starch, yeast extract or starch and yeast extract
This
is in good agreement with the result reported by Komastsu et al. (1991) which suggested
that saturation of unsaturated LCFA in acidogenic reactor was one of the main reasons why
two-phase anaerobic system could reduce the toxicity of LCFA. However, the ratio of
palmitic acid was 6% and 8% under mesophilic and thermophilic conditions when fed only
oil, suggesting that the addition of starch and yeast extract stimulated the saturation of
unsaturated LCFA in acidogenic reactor.
8.3.3 Microbial population shifts under different feed composition, temperature and
HRT in acidogenic reactor. The LogC value of the numbers of Bacteria cells was
between 11 and 12 in the inoculum used in the experiments.
At the mesophilic
temperature, the numbers of Bacteria cells were lowest when only canola oil was fed and
decreased with the increase of HRT (Table 8.2), implying that canola oil inhibited the
growth of Bacteria cells. When fed with starch, yeast extract or starch and yeast extract,
the bacteria numbers increased with the increase of HRT, suggesting that longer HRT
benefited the growth of bacteria under co-digestion situation. The population levels when
canola oil and starch were fed were higher than that when the canola oil and yeast
extracts were fed at every HRT, indicating that starch had the better effects on enhancing
153
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
the growth of Bacteria cells under the inhibition of canola oil. The numbers of Bacteria
cells when canola oil, starch and yeast extract were fed was the highest at all HRT. This
suggested that the combination of canola oil, starch and yeast extracts had the most
effective results on increasing the Bacteria population. Even though the highest bacterial
population was achieved under feeding of oil, starch and yeast extract, the extent of oil
hydrolysis under this condition was lower than that when oil and starch were fed
(Fig.8.1a), implying that the hydrolysis of oil was not only related to the total population
number, it might also related to the numbers of bacterial subpopulations in the acidogenic
reactor.
HRT
1 day
2 days
3 days
4 days
5 days
6.3
6.1
6
5.4
5.7
oil
thermophilic 3.5
4.8
4.8
5.1
5.8
oil
mesophilic
9.3
10.2
11.2
11.3
starch
thermophilic 10.3
11.2
11
10.8
10.6
oil
mesophilic
6.1
9.2
9.1
9.3
10.3
protein
thermophilic 9.6
9.6
10.2
9.8
8.9
9.2
11.2
11.3
12.1
12.3
9.9
9.7
9.5
mesophilic
oil, starch mesophilic
protein
7.2
7.1
thermophilic 10.4
Table 8.2 Bacteria population (LogC of cell counts. cm -2) on varied organic components,
temperature and HRT.
At thermophilic temperature, the shifts of the number of Bacteria cells varied significantly
with that at mesophilic temperature. When only fed with oil, the Bacteria population
154
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
increased with the increase of HRT.
There was no significant change of the number of
Bacteria cells in different HRT when fed with oil and starch or oil and yeast extract,
indicating that the number of Bacteria cells was independent of HRT at these two feeding
compositions. When fed with oil, starch and yeast extract, the number of Bacteria cells
increased up to HRT of 2 days, it decreased when HRT was greater than 2 days.
The LogC value of numbers of Archaea cells in original inoculum was about 8 to 9. Table
8.3 shows that the number of Archaea cells were very low both in mesophilic and
thermophilic conditions and had no difference at various HRT when only fed with oil.
However, when fed with starch, yeast extract or starch and yeast extract, the number of
Archaea cells all increased compared to that when only fed with oil, suggesting that Archaea
cells could grow effectively when fed oil with starch, yeast extract or both. This
demonstrated that the combination of organic materials is essential for the growth of
Archaea cells. When fed by oil and starch or oil and yeast extract, the number of Archaea
cells all increased with increase of HRT both at mesophilic and thermophilic temperature.
When fed by oil, starch and yeast extract, Archaea population increased with the increase of
HRT at mesophilic temperature. However, at thermophilic temperature, it increased with the
increase of HRT up to 3 days and started decreasing when HRT>3days.
155
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
HRT
mesophilic
1 day
2 days
3 days
4 days
5 days
2.1
2.2
2.3
2.1
2.2
oil
thermophilic 2
2.2
2
2
2.3
oil
mesophilic
4.1
7.3
7.4
8.4
starch
thermophilic 3.6
6.9
7.2
7.4
7.8
oil
mesophilic
2.3
3.2
5
5.1
5.7
yeast extract
thermophilic 2.6
5.5
6.3
7.2
7.3
oil, starch
mesophilic
2.2
4.3
7.1
8.4
9.3
yeast extract
thermophilic 2.8
6.2
8.1
5.7
5.3
2
Table 8.3 Archaea population (LogC cell counts. cm -2 ) on different organic components,
temperature and HRT.
Zeeman and Sanders (2001) suggested that Archaea cells were very important to the lipid
hydrolysis and because not too many Archaea cells existed in acidogenic reactor, two-phase
will not benefit lipid hydrolysis. This experiment showed that at HRT of 1 day, the numbers
of Archaea cells were low and had no significant difference at mesophilic and thermophilic
temperature when fed with different feed compositions. However, the extent of oil
hydrolysis in different feed compositions and temperature were different, implying that the
lipid hydrolysis efficiencies may not be related to the number of Archaea cells. The reason
might be related to the facts that with the higher Archaea population, VFA could be
consumed fast and this will enhance the acidogenesis, acetogenesis and hydrolysis process.
156
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
8.3.4 The hydrolysis of canola oil in methanogenic reactor.
The extent of oil
hydrolysis in methanogenic reactor was estimated as the oil concentration of the effluent
of methanogenic reactor subtracted from the oil concentration of effluent from acidogenic
reactor then divided by oil concentration of effluent from acidogenic reactor. The total
extent of oil hydrolysis in the two phase system was calculated by subtracting the oil in
the effluent of reactor from total oil fed into the acidogenic reactor and divided by the
total oil fed into the acidogenic reactor. The data showed that the extent of oil hydrolysis
in methanogenic reactor was very low when only fed oil at both mesophilic and
thermophilic temperature. The extent of oil hydrolysis increased with the increase of
HRT at all feed compositions, suggesting that longer HRT benefited the hydrolysis of oil
in the reactor. At thermophilic temperature, the extent of oil hydrolysis was highest when
fed by oil, starch and yeast extract, while feeding with oil and starch achieved highest
extent of oil hydrolysis at mesophilic temperature. This was different with the oil
hydrolysis in acidogenic reactor in which feeding oil and starch could achieve highest
extent of oil hydrolysis both at mesophilic and thermophilic temperature.
However, contradicting these findings, Zeeman and Sanders (2001) observed that lipid
hydrolysis cannot proceed in the absence of methanogenesis and the two-phase system
will not improve the efficiencies of lipid hydrolysis.
In this experiment, with co-
digesting with starch, yeast extract or starch and yeast extract the total oil hydrolysis in
two-phase UASB anaerobic systems as all higher than 80% and it was much greater than
that in the one-phase UASB (below 40%) as shown in Chapter 7.
157
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
Oil hydrolysis in
Total oil hydrolysis
methanogenic reactor (%)
in two reactors (%)
Mesophilic
Thermophilic
Mesophilic Thermophilic
1
4.2
8.6
5.3
12.4
2
5.5
8.0
7.7
12.7
3
5.6
8.8
7.9
12.6
4
4.8
9.5
8.5
13.4
5
5.2
9.1
8.2
12.5
1
83.7
83.3
87.1
93.2
2
84.3
84.3
88.6
93.7
3
85.3
84.7
89.3
94.3
4
85.7
85.4
87.7
95.2
5
87.9
85.2
91.8
95
1
75.3
80.5
80.2
87.6
+yeast extract 2
75.5
80.7
80.3
88.4
3
75.3
80.8
80.8
88.7
4
75.9
80.6
81.4
88.3
5
76.2
81.8
81.2
90.8
1
76.8
87.2
82.5
88.2
+yeast extract 2
77.9
87.8
83
90.4
3
79.3
88.2
83.6
92.2
4
80.7
89.8
84.9
94.3
5
83.8
89.9
89.8
94.6
Feed
HRT
substrate
Oil
Oil+starch
Oil
Oil+starch
Table 8.4. Extent of oil hydrolysis at various HRT in methanogenic reactor.
158
Metahne gas
production (ml/l.day)
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
800
a
600
oil+starch
400
oil+YE
oil+YE+starch
200
0
1
2
3
4
5
Metahne gas
production (ml/l.day)
HRT (days)
800
b
600
oil+starch
400
oil+YE
oil+YE+starch
200
0
1
2
3
4
5
HRT (days)
Fig.8.6 Methane gas production in different HRT at (a) mesophilic and (b) thermophilic
temperature.
HRT
mesophilic
1 day
2 days
3 days
4 days
5 days
0
0
0
0
0
oil
thermophilic 0
0
0
0
0
oil
starch
mesophilic
43.9
42.8
43.1
45
thermophilic 45.9
52
54.8
55.8
55.7
oil
protein
mesophilic
45.0
thermophilic 45.1
mesophilic
42.8
49.2
55.7
49
46.9
43.6
50.2
53.1
53.7
51.5
54.8
56.7
50.9
thermophilic 45.9
55.7
59
52.5
55.7
oil,
starch
protein
44.6
Table 8.5. The extent of oil methanogenesis at various HRT in the methanogenic reactor.
159
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
When the reactor was only fed with oil, no methane gas was produced both at mesophilic
and thermophilic temperature. Figure 8.6 shows that at mesophilic temperature, the methane
gas production was highest at all HRT when fed with oil, starch and yeast extract, while the
methane gas production was high when fed with oil and yeast extract than oil and starch.
When HRT >1 days, the difference of the methane gas production at mesophilic and
thermophilic was not significant, suggesting that thermophilic temperature was not required
for methanogenic reactor when co-digesting oil with starch or yeast extract.
The rate of methanogenesis was estimated as the methane gas produced at different feed
compositions divided by the theoretical total methane gas that could be produced by feeding
2 gCOD/l at different HRTs. For example, when fed with oil and starch and HRT of 1 day
at mesophilic temperature, 624 ml methane was produced from the methanogenic reactor,
the theoretical amount of methane gas that could be produced from methanogenic reactor
was 1400 ml (350 ml/gCOD multipled by 4 g COD), thus 44.6% methanogenesis was
achieved.
From the data show in Table 8.5, the methanogenesis at mesophilic and
thermophilic temperature were around 45 to 55%, suggesting that mesophilic and
thermophilic had no significant on methanogenesis in methanogenic reactor of two-phase
system and that complete mineralisation of feed to methane was not achieved. One gCOD/l
of oil feed might be too high, thus, a detailed study of effects of oil concentration on twophase UASB anaerobic process was performed and is discussed in Chapter 9.
160
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
8.4 Summary and conclusion
The results of this study showed that co-digestion oil with starch and yeast extract was
effective on enhancing the oil hydrolysis in two-phase UASB anaerobic process. One
reason might be related to the fact that the co-digestion stimulated that growth of Archaea
and Bacteria cells under inhibition of oil or LCFAs which was hydrolysed from oil.
Another reason might be attributed to the fact that co-digestion enhanced the significant
saturation of unsaturated LCFA in acidogenic reactor.
The extent of starch and yeast extract hydrolysis decreased when co-digested with canola oil
at both thermophilic and mesophilic temperature. This might be related to the fact that oil
inhibited the growth of both Archaea and Bacteria cells. Interestingly, in combination with
oil or with oil and yeast extract, the amount of starch hydrolysed at mesophilic temperature
was higher than at thermophilic temperature, while the extent of yeast extract hydrolysed
was lower at mesophilic temperature than at thermophilic temperature when fed together
with oil or oil and starch. At mesophilic temperature, the amount of starch hydrolysed
increased when HRT increased from 1 to 2 days under oil or oil and yeast extract feeding.
At thermophilic temperature, the extent of starch hydrolysis decreased with the increase of
HRT, while the extent of yeast extract was relatively independent of HRT.
In acidogenic reactor, there was no methanogenesis to be achieved and the extent of oil
hydrolysis when fed by oil, oil and starch, oil and yeast extract or oil, starch and yeast
extract was higher at thermophilic temperature than at mesophilic temperature. Furthermore,
161
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 8 Lipid hydrolysis in mesophilic and thermophilic two-phase anaerobic treatment systems.
the addition of starch was most effective in the oil hydrolysis compared to the addition of
yeast extract or starch and yeast extract. The extent of hydrolysis of the oil was independent
of HRT when HRT>2days under all feed conditions.
In the methanogenic reactor, high oil hydrolysis was also achieved accompanied with the
methanogenesis under co-digestion with starch, yeast extract or starch and yeast extract.
More than 80% oil was hydrolysed after the acidogenic and methanogenic reactors,
suggesting that two-phase UASB anaerobic systems have the significant advantage on lipid
hydrolysis which is currently the most difficult step in wastewater treatment plant.
However, the methanogenesis in the methanogenic reactor was not significant.
162
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating
bacteria growth with co-digestion of proteins and carbohydrates in twophase UASB reactor
9.1 Introduction
Unprocessed urban domestic wastewater carries about 40-100 mg lipids/l (Quemeneur and
Marty, 1994) which mainly come from kitchen water and human excreta. In industrial
wastewater, lipid concentrations vary, for example, slaughterhouses wastewater could
contain 350-520 mg/l lipids and wool scouring wastewater could contain 2 –15 g/l (Rinzema,
1988). Lipid could be successfully degraded by aerobic processes, however, even very low
lipid loading rate (1-2 g COD/l·day) resulted in wash-out of granular sludge in an anaerobic
expanded granular sludge bed (EGSB) reactor (Rinzema, 1988). Two-phase process is
designed to achieve optimum growth of acidogens in acidogenic reactor and methanogens in
methanogenic reactor. However, the continuous and constant organic overloaded result in
process failure (Fongastitkul and Lo, 1994). Although conventional two-phase anaerobic
digesters have been used for sewage treatment, these systems still face problem of relatively
poor performance of the methanogenic reactor (Fongastitkul and Lo, 1994). The data of
Chapter 8 also showed that even though high extent of hydrolysis could be achieved at a
loading rate of emulsified canola oil of 1 g COD/l.day, the methanogenesis of the oil in the
methanogenic reactor was poor (30 to 50%) in a two-phase UASB reactor system with codigestion of starch and yeast extract.
163
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
In recent years, a number of large-scale biogas plants have been built to treat organic
industrial waste from slaughterhouse, food processing industries and animal manure which
are rich in lipids, protein and carbohydrates. These plants have achieved very good results
(Ahring et al., 1992; Edelmann et al., 2000). It was difficult to digest wastewater only rich
in lipid or lipid and protein. Carbohydrate was necessary in the digestion of lipid and protein
rich wastewater (personal communication with operator in Spearwood Wastewater treatment
Plant). The Chapter 8 showed that the addition of carbohydrate produced a better result than
protein on enhancing the extent of hydrolysis of lipid.
Co-digestion of lipid with
carbohydrate and protein could decrease the inhibition of lipid on anaerobic digestion and
achieve greater acidogenesis and methanogenesis than when only fed with lipid (Chapters 7
and 8). However, the effects of concentration or ratio of lipid, carbohydrate and protein on
lipid anaerobic digestion is still unclear. The aim of this study is to investigate the effects of
concentrations of protein and carbohydrate on the anaerobic degradation of lipid, as well as
on the population shifts of anaerobic bacteria communities during two-phase anaerobic codigestion. Chapter 8 showed that higher extent of oil hydrolysis could be achieved at
thermophilic temperature than at mesophilic temperature in the acidogenic reactor, while the
extent of oil methanogenesis was not significantly different at mesophilic and thermophilic
temperature in two-phase UASB reactor. Thus, in this experiment, thermophilic temperature
was chosen for acidogenic reactor while mesophilic temperature was chosen for
methanogenic reactor.
164
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
9.2 Materials and methods
9.2.1 Experimental reactor set up. One, two-phase UASB reactor was set up as described in
section 3.1.2. The acidogenic reactor was kept at 55°C while the methanogenic reactor was
kept at 37°C Inoculation was performed as described in the section 8.2.1. After inoculation,
the reactor was fed 1.5 g COD/l of starch at hydraulic retention time (HRT) of 1 day for 14
days until more than 80% COD reduction had been achieved. After reactivation, the reactors
were fed 500 mgCOD/l canola oil for 14 days as the control. Then the system was restarted
by inoculating as before and feeding 1.5 g COD/l of starch for 14 days until more than 80%
COD reduction was achieved.
The system was fed a mixture of lipid (canola oil),
carbohydrate (starch) and protein (yeast extract) at various concentrations as shown in the
Table1 at hydraulic retention time (HRT) of 1 day.
Eight experiments were carried out to examine the effects of the addition of protein and
carbohydrate on the anaerobic degradation of lipid. Each run lasted 30 days. The lipid
degradability and microbial population shifts in the methanogenic reactor were investigated.
The solutions of three organic materials were freshly prepared and autoclaved before the
experiment and stored in the refrigerator at 4 0C to avoid the degradation by contamination.
The feed solution was stirred continuously. HRT for acidogenic reactor and methanogenic
reactor was controlled at 24 hours. The pH in the acidogenic reactor was controlled at 5 to 6
by addition of 10 N NaOH and 2M HCl.
165
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
Component
Canola oil (mgCOD/l)
Starch (mgCOD/l)
Yeast extract
(mgCOD/l)
Control
500
0
0
Run1
500
1000
500
Run2
500
1500
500
Run3
500
2000
500
Run4
500
2500
500
Run5
500
500
1000
Run6
500
500
1500
Run7
500
500
2000
Run8
500
500
2500
Table 9.1. Composition of feed in the different runs.
1 g carbohydrates (starch) is equivalent to 1.01 g COD; 1 g lipid (canola oil) is equivalent to 2.86 g COD; 1 g
protein (yeast extract) is equivalent to 1.35 g COD
9.2.2 Sampling and analysis. Procedure for sampling and analysis of microbial
communities was as described in section 3.4. Procedure for measurement of canola oil,
yeast extract and starch was as described in section 3.7. Analysis of LCFA in acidogenic and
methanogenic reactor was as described in section 3.7.1.
166
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
9.3 Results and discussion
9.3.1 The inhibition of oil on acidogenic and methanogenic reactors.
In the control experiment which was fed only oil, there was no methane produced in
acidogenic reactor and the pH was about 6.3-6.9. The extents of hydrolysis and degradation
of canola oil in both acidogenic reactor and methanogenic reactor were lower than 20% after
two days (Fig. 9.1), indicating that canola oil inhibited the acidogenesis and methanogenesis
even at low concentration (500 mgCOD/l) in the feed. This was in good agreement with the
reports in the literature (Rinzema, 1988; Angelidaki et al., 1990) and previous experiments
(Chapters 7 and 8).
Interestingly, the extent of oil hydrolysis at 1000 mg COD/l
concentration of oil at HRT of 1 day (Fig.8.1b) in acidogenic reactor at thermophilic
temperature was about 5% and it was similar with that at 500 mg COD/L (Fig.9.1) at day 14.
Therefore the extent of oil hydrolysis in acidogenic reactor did not increase with the
decrease of oil concentration. The hydrolysis of oil in the methanogenic reactor was also
very low, this was also in accordance with the data shown in Table 8.4 in which the extent
of oil hydrolysis was about 5 % in the methanogenic reactor at mesophilic temperature. The
experiment demonstrated that oil was difficult to hydrolyse even at low lipid loading in the
two-phase anaerobic reactor.
167
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
50
45
40
35
30
25
20
15
10
5
0
Acidogenic
Methanogenic
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Days
Fig. 9.1. Degradation of canola oil in the acidogenic and methanogenic reactors.
900
800
700
600
Series1
500
400
300
200
100
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Days
Fig.9.2 Methane production from methanogenic reactor after feeding with canola oil.
The methane gas production decreased upon feeding 500 mg COD/l of canola oil in the
acidogenic reactor (Fig. 9.2). The methanogenic activity decreased to zero after 4 days
168
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
feeding canola oil as indicated by zero-methane production from day 4 onward. From day 3,
the sludge in the methanogenic reactor started to wash out. This might be related to the
adsorption of lipid on sludge as discussed by Hwu (1997).
The results from Fig.9.1 and Fig.9.2 suggest that in the absence of other organic component
in feed, even low concentration of oil (500 mgCOD/l) would inhibit the acidogenesis and
methanogenesis in the two-phase anaerobic process.
9.3.2 The influence of organic composition on the oil degradation
Figure 9.3 shows the methanogenic activity when the reactors were fed with different
compositions of starch and yeast extract with 500 mgCOD/l of oil. Assuming that all the
COD fed was degradable, the extent of methanogenesis was calculated as the mean value of
the daily methane gas production during the different runs divided by the theoretical
expected daily methane production rate, calculated as 350 ml/gCOD in fed (shown at Table
9.2). For example, when the reactor was fed with 2 g COD/l total COD at oil: starch: yeast
extract ratio of 1:2:1 in COD weight (Run1) at HRT of 1 day, 1400 ml gas was theoretically
assumed to be produced from the reactor. However, only 655 ml of methane was produced.
Thus the extent of methanogenesis was calculated as 655 ml divided by 1400 ml i.e. 47%.
With the increase of starch amount, the methane production increased, implying that starch
benefited the methanogenesis. This is in accordance with results of previous experiments
(Hanaki et al., 1987) which reported that carbohydrates were easily converted to acid during
the complex organic material anaerobic degradation. However, when the concentration of
169
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
yeast extract increased from 500 to 2500 mgCOD/l, the extent of methane production
decreased significantly, suggesting that yeast extract was not as effective as starch in
methanogenesis process.
With the increase of yeast extract feeding, the extent of
methanogenesis decreased, implying that the degradability of yeast extract was not as high
as that of starch.
This is also in agreement with the experimental results achieved
previously, which suggested that protein was difficult to be degraded due to its lower extent
of hydrolysis (Hanaki et al., 1987; Batstone, 1999). These results suggested that the higher
carbohydrate concentration or ratio in anaerobic co-digestion of complex organic waste
treatment would benefit the methanogenesis of the two-phase anaerobic process.
The extent of methanogenesis was the highest when COD weight ratio of Oil:Starch:Yeast
extract was 1:5:1, suggesting that higher starch concentration in co-digestion oil, starch and
yeast extract was essential. The results indicated that there was an optimal composition of
protein, lipid and carbohydrate for methanogenesis although it might be different in different
anaerobic reactors. The extents of methanogenesis from Run 1 to Run 4 were all higher than
those fed with 1000 mgCOD/l.day of oil as shown in Fig.8.7, indicating that lower oil
concentration also benefit the methanogenesis in the two-phase UASB anaerobic system.
170
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
Run
1
2
3
4
5
6
7
8
Oil:Starch:Yeast
1:2:1
1:3:1
1:4:1
1:5:1
1:1:2
1:1:3
1:1:4
1:1:5
47.3
68.5
71.6
80.7
37.2
36.5
35.7
30.
extract (COD g)
Percentage of
methanogenesis
Table 9.2 The extent of methanogenesis in methanogenic reactor for different feed
compositions.
171
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion of proteins and carbohydrates in two-phase UASB
reactor
Methane production (ml/day)
2000
Run1
1800
Run2
Run3
Run4
Run5
Run7
Run8
Run6
1600
1400
1200
1000
800
600
400
200
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Time (days)
Fig. 9.3. Effect of different ratio of organic composition on methane production in methanogenic reactor.
172
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion of proteins and carbohydrates in two-phase UASB
reactor
Percentage of degradation
100
90
80
70
60
50
40
30
Run1
Run2
Run3
Run4
Run5
Run6
Run7
Run8
20
10
0
0
10
20
30
40
50
60
70
80
90
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240
oil
Time (days)
starch
yeast extract
Fig. 9.4. Extent of oil, starch and yeast extract degradation in methanogenic reactor with different feed compositions
173
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion of proteins and carbohydrates in two-phase UASB
reactor
50
Run1
Run2
Run3
Run4
Run5
Run6
Run7
LCFA concentration (mg/l)
40
Run8
30
20
10
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Time (days)
Fig.9.5. The concentration of LCFA in the effluent of methanogenic reactor under different feeding modes
174
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion
of proteins and carbohydrates in two-phase UASB reactor
Figure 9.4 shows a comparison of the final breakdowns of starch, lipid and yeast extract
after the second reactor under the different feeding modes. With the increase of starch
addition from 1000 to 2500 mg COD/l.day (Run 1-Run 4), the lipid degradation increased
from 71 to 92%. With increase of protein addition (Run5- Run8) from 1000 to 2500 mg
COD/l.day, the degradation of lipid decreased from 82 to 65%. It suggested that higher
starch was more effective than higher yeast extract in enhancing the oil degradation.
With the increase the loading rate of starch, the degradation of starch was kept around 85%.
However, with the increase of yeast extract, the degradation of starch decreased from 85 to
75%, suggesting that higher yeast extract could also inhibit the degradation of starch. The
degradation of yeast extract in the first four Runs was around 55%, and it decreased from
55% to 36% with the increase of yeast extract from 1000 to 2500 mg COD/l.day.
The concentration of LCFA in the effluent of methanogenic reactor decreased from 25 mg/l
to 7 mg/l with the increase of starch concentration, suggesting that the addition of starch
benefited the LCFA transformation. With the increase of yeast extract feeding, the LCFA
concentration increased, indicating that higher yeast extract concentration or lower starch
concentration inhibited the LCFA transformation. With the higher LCFA concentration
accumulation, the methane production decreased and was kept lower from Run 5 to Run 8.
This might explain the result showed in the Chapter 8 in which even though higher
hydrolysis of oil was achieved the extent of methanogenesis was low.
175
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with co-digestion of proteins and carbohydrates in two-phase UASB
reactor
14
13
Cell numbers (LogC)
12
11
10
9
8
7
Run1
6
Run2
Run3
Run4
Run5
Run6
Run7
5
Archaea
4
Bacteria
3
SRB
Run8
2
1
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
Time (days)
Fig.9.6 Archaea, Bacteria and SRB shifts under the different feeding compositions in methanogenic reactor.
176
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with codigestion of proteins and carbohydrates in two-phase UASB reactor
9.3.3 The effects of addition of different composition substrates on the shifts of
microbial subpopulations in methanogenic reactor
Figure 9.6 illustrates the results obtained with oligonucleotide probes ARC 915
(targeting Archaea cells), EUB 338 (targeted to Bacteria cells including mainly
acetogens and acidogens in the reactor) and SRB (targeting to sulfate-reducing bacteria).
After reactivating by feeding 1.5 g COD/l carbohydrate, the number of Archaea cells
was higher than the number of Eubacteria cells and the numbers of SRB cells. After
feeding with oil, starch and yeast extract, the numbers of Archaea cells decreased sharply
from LogC value of 13 to 10 and became lower than the numbers of Bacteria cells,
implying that the growth of methanogens was more sensitive to lipid than acetogens and
acidogens. With the increase of starch concentration, the numbers of Archaea cells
increased from LogC value of 10.2 to 13.8, indicating that starch benefited the recovery
of growth of Archaea cells. This result could be related to the increase of methanogenic
activity and lipid degradation in the methanogenic reactor with the increase of starch
feeding. Contrary to the expectation, the numbers of Bacteria and SRB cells were
relatively stable with the increase of starch, indicating that increasing of addition of
starch did not affect the population of Eubacteria cells in the methanogenic reactor.
With the increase of addition of yeast extract from Run 5 to Run 8, the numbers of
Archaea cells decreased from LogC value of 13.8 to 10.4, while both the numbers of
Bacteria cells and SRB cells increased from 11.4 to 13.6 and 8.2 to 12.3, respectively.
Higher LCFA concentration accumulated from Run 5 to Run 8, and higher LCFA
inhibited the growth of Archaea cells. Furthermore, some of the amino acids hydrolysed
from yeast extract, like methaionine and cysteine contain sulphur, therefore, the increase
177
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with codigestion of proteins and carbohydrates in two-phase UASB reactor
of addition of yeast extract resulted an increase of the sulphur concentration which
benefits the growth of SRB. The competition between SRB cells and Archaea cells
reported in previous experiments (Raskin et al., 1995; Raskin et al., 1996) showed that
the increase of sulphur would benefit the competition of SRB. This experiment showed
that Bacteria and SRB cells were more competitive than Archaea cells in a protein
abundant environment. Beside substrate competition, it was reported that SRB and
methanogens are also competitive on hydrogen (Lovley at al., 1982).
With the decrease of Archaea cells, the LCFA accumulated in the reactor, implying that
LCFA transformation was indirectly related to the numbers of Archaea cells, although
some other parameters, such as hydrogen pressure might indirectly relate to the Archaea
cells. The Chapter 8 showed that lipid hydrolysis was not related to the numbers of
Archaea cells, however this chapter shows that LCFA transformation or degradation
depend on Archaea cells and the transformation of LCFA might be more effective with
more Archaea cells.
The microbial sub-population shifts observed under the different feeding conditions
provided some more information that could not be illustrated by monitoring the physicochemical parameters. Koster and Cramer (1986) suggested that the inhibition of lipid
and toxicity of LCFA was affected by the concentration of lipid or LCFA, and not
related to the numbers of anaerobic bacteria. This experiment obviously showed that the
numbers of different microbial communities were related to the degradation levels of
lipid or LCFA. The microbial investigation during the anaerobic digestion process is
therefore essential for the control of the digester performance. The success of codigestion of lipids, protein and carbohydrates is explained by the fact that these three
178
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 9. Enhancing the lipids anaerobic digestion by stimulating bacteria growth with codigestion of proteins and carbohydrates in two-phase UASB reactor
organic materials are all necessary in the growth of anaerobic bacteria and providing
these nutrients will benefit the microorganisms. The different composition of three
organic components could stimulate the growth of different bacterial communities.
Microbial shift investigation is very suitable indicator to design or explain the optimal
co-digestion composition of lipid, protein and carbohydrate.
9.4 Summary and Conclusion
In this chapter, the influence of concentrations of starch and yeast extract on canola oil
degradation in two-phase UASB anaerobic systems was evaluated. The results showed
that the concentration of starch higher than that of oil and yeast extract in co-digestion
could enhance the degradation of oil. Higher starch feeding probably enhanced the
LCFA transformation, while higher yeast extract loading would decreased the
degradation of oil.
The microbial population study present more detailed information to explain the
experimental results. The addition of higher starch concentrations stimulated the growth
of Bacteria cells thus positively increase the Archaea cells number, while higher yeast
extract concentrations also benefit the growth of Bacteria and SRB cells. The later cells
competed with the growth of Archaea cells, thus decreased the methanogenesis in
methanogenic reactor. The microbial data also showed that LCFA transformation or
degradation might be related to the numbers of Archaea cells as the LCFA accumulation
increased with the decrease of Archaea cell numbers.
179
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 10. Conclusion and recommendation
Chapter 10. Conclusion and recommendation
10.1 Conclusion
In theory, lipids and their hydrolysis products long chain fatty acids (LCFA) are readily
biodegradable.
However, two main difficulties have to be overcome before achieving
efficient and reliable anaerobic treatment of lipids. First is the inhibition of LCFAs on
methanogenic and acetogenic bacteria. Second is the wash-out of the biomass due to the
flotation resulting from the adsorption of lipids. Some efforts have been done in developing
new reactor designs and loading rate control. However, the effect of influent substrate study
seems very limited. Recently, combined biodegradation of complex substrate or full scale
co-digestion of organic waste has attracted more and more interest (Ahring et al. , 1992;
Kubler et al., 2000).
It has been noticed that the co-digestion of lipids, protein and
carbohydrates would benefit organic wastewater treatment. However, the study of organic
co-digestion only in the primary stage due to the fact that co-digestion is a complex process.
The relation between combination of organic substrates that are fed and efficiency of lipid
hydrolysis is still unclear. Most of the difficulty in the study of co-digestion is the
investigation of microbial communities shifts under the different combination of organic
substrates. Monitoring or controlling of this process is correctly conducted using physicochemical parameters, such as VFA, methane volume, pH, temperature, COD, loading rate
and HRT. Although the microbial control of co-digestion process is also a very important
aspect, the research of this field is scarce.
180
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 10. Conclusion and recommendation
This study applied microbial techniques as well as physico-chemical parameters to
investigate the effects of the combination of substrates on the reduction of toxicity of LCFA
on anaerobic bacteria and the enhancement of the anaerobic degradation of lipid. The
experimental results can be divided into two groups:
Chapters 4,5 and 6 dealt with the research of the influence of glucose and amino acid on the
toxicity of LCFA on anaerobic bacterial groups, as well as on the granule formation under
LCFA and on the recovery from inhibition of LCFA. It was proven in the laboratory scale
that glucose and amino acid were effective in decreasing the toxicity of LCFA on the
anaerobic bacteria, thus enhancing the degradation of LCFA. Although glucose gave better
results than cysteine on reducing the toxicity of LCFA and recovering the regrowth of
anaerobic bacteria, the combination of glucose and cysteine was optimal for the reduction of
toxicity of LCFA.
Chapters 7, 8 and 9 were directed to investigate the influence of carbohydrates and protein
on the hydrolysis and anaerobic degradation of lipids. The results showed that the integrated
addition of protein and carbohydrate with lipid had the optimal effect on the hydrolysis of
lipid compared to the separate addition of protein or carbohydrate. The main reason might
be that the combined addition of protein and carbohydrate benefited the growth of bacteria
and thus decreased the loss of biomass. The co-digestion of protein, carbohydrate and lipid
decreased the loss of microbial population and thus enhanced the digestion of lipids.
181
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 10. Conclusion and recommendation
Based on the observations and results of this thesis, the following conclusions could be
drawn:
•
The addition of glucose or amino acid could reduce the toxicity of LCFA on anaerobic
bacteria, thus benefit the anaerobic degradation of LCFA
•
The presence of glucose could benefit the growth of methanogens, acidogens and
acetogens, thus enhance the granule formation under the inhibition of LCFA.
•
When the reactor was shut down and anaerobic bacteria were inhibited by LCFA, glucose
and cysteine could enhance the recovery of anaerobic bacteria growth.
•
The co-digestion of carbohydrate, protein and lipid benefited the hydrolysis of lipid and
then increased the anaerobic degradation of lipid.
•
Two stage digestion was more effective than one stage digestion in lipid hydrolysis but
presented low methanogenesis at high lipid loading rate. Lipid hydrolysis occurred both in
acidogenic and methanogenic reactors.
•
Biological monitoring of lipid degradation combined with other organic materials,
carbohydrate and protein in anaerobic digester was necessary and could avoid some failure
resulting from wrong operation by early warning and detailed inside microbial information.
It should be included in the routine analyses performed in the wastewater treatment plants.
182
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 10. Conclusion and recommendation
•
The co-digestion of protein and carbohydrate enhanced the anaerobic degradation of lipid
through the stimulating the growth of anaerobic bacteria subpopulation under the inhibition
of lipid.
•
While starch was more effective in stimulating the growth of Archaea cells, yeast extract
more benefited the growth of Bacteria cells. The number of Archaea cells might be related
to the LCFA transformation during the lipid degradation.
10.2 Recommendation for future works
In this study, cysteine and sodium oleate were chosen to represent the amino acids and
LCFA, canola oil and yeast extract to represent lipid and protein. Before the practical
application of co-digestion of substrate and lipids or LCFAs, more products need to be
tested.
The relation between the combination ratio among the lipid, protein and
carbohydrate and growth rate of anaerobic bacteria also need further examination and more
detailed microbial and physico-chemical modelling, computer software designing before
practical application.
Because this study is limited to lab-scale experiments, the
investigation of integrated addition or co-digestion in the practical wastewater treatment
plants is still needed.
Although the physico-chemical as well as microbial parameters have been investigated
during the co-digestion of lipid with starch and yeast extract, or LCFA with glucose and
cysteine, the pathway and co-metabolism of carbohydrates and protein on enhancing the
183
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
Chapter 10. Conclusion and recommendation
lipid degradation, or glucose and amino-acids on stimulating LCFA transformation still have
not been clarified. Thus, future works on these fields are still necessary.
184
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
REFERENCE
Ahring, B.K., Sandberg, M. and Angelidaki, I. (1995). “Volatile fatty acids as indicators of process
imbalance in anaerobic digesters.” Appl. Microbiol. Biotechnol. 43: 559-565.
Ahring, B.K., Angelidaki, I. and Johansen, K. (1992). “Anaerobic treatment of manure together
with industrial waste.” Wat. Sci. Tech. 25: 311-318.
Alexander, M. (1996) "Biodegradation of organic chemicals." Environ. Sci. Technol. 18: 106-111
Alphennaar, P.A., Groeneveld, N. and Van Aelst, A.C. (1994). “Scanning electron Microscopical
method for internal structure analysis of anaerobic granular sludge.” Micron. 25: 129-133.
Amann, R., Krumholz, L. and Stahl, D. (1990a). “Fluorescent-oligonucleotide probing of whole
cells for determinative, phylogenetic, and environmental studies in microbiology.” J. Bacteriol.
172(2): 762-770.
Amann, R.I. Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R. and Stahl, A. (1990b).
“Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analysing
mixed microbial population.” Appl. Environ. Microbiol. 56(6): 1919-1925.
Amann, R. I., Stromlet, J., Devereux, R. Key, R. and Stahl, D.A. (1992). “Molecular and
microscopic identification of sulfate-reducing bacteria in multispecies biofilms.” Appl. Environ.
Microbiol. Feb.: 614-623.
185
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Angelidaki, I., Petersen, S.P. and Ahring, B.K. (1990). “Effects of lipids on thermophilic anaerobic
digestion and reduction of lipid inhibition upon addition of bentonite.” Appl. Microbiol.
Biotechnol 33: 469-472.
Angelidaki, I and Ahring, B.K. (1995). “Establishment and characetrization of an anaerobic
thermophilic (550 C) enrichment culture degrading long-chain fatty acids.” Appl. Environ.
Microbiol. 61(6): 2442-2445.
Angelidaki, I., Ellegaard, L. and Ahring, B.K. (1998). “A comprehensive model of anaerobic
bioconversion of complex substrates to biogas.” Biotech. Bioeng. 63(3): 363-372.
APHA (1985) "Standard Methods for the Examination of Water and Wastewater", American
Public Health Association, New York
Archer, D.B. (1984). “Detestations and quantification of methanogens by enzyme-linked
immunosorbent assay.” Appl. Environ. Microbiol. 48: 797-801.
Balch, W.E., Fox, G.E., Magrum, L.J. Woese, C.R. and Wolfe, R.S. (1979). “Methanogens:
revaluation of a unique biological group.” Microbiol. Rev. 43(2): 260-296.
Batstone, D. J. (2000), "High rate anaerobic treatment of complex wastewater". PhD thesis,
Queensland University, Australia.
186
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Beccari, M., Bonemazzi, F., Majone, M. and Riccardi, C. (1996). “Interaction between
acidogenesis and methanogenesis in the anaerobic treatment of olive oil mill effluents.” Wat. Res.
30(1): 183-189.
Beccari, M., Majone, M. and Torrisi, L. (1998). “Two-reactor system with partial phase separation
anaerobic treatment of olive oil mill effluents.” Wat. Sci.Tech. 38(4-5): 53-60.
Boone, B.Z. and Bryant, M.P. (1980). “Propionate-degrading bacterium, Sytrophobacter wolinii
sp.nov.gen.nov., from methanogenic ecosystems.” Appl. Environ. Microbiol. 40: 626-632.
Boone, D.R., Johnson, R.L. and Liu, Y. (1989). “Diffusion of the interspecies electron carriers H2
and formate in methanogenic ecosystems and its implications in the measurement of Km for H2 or
formate uptake.” Appl. Environ. Microbiol. 55: 1735-1741.
Britz, T.J., Spangenberg, G. and Venter, C.A. (1994). “Acidogenic microbial species diversity in
aerobic digesters treating different substrates.” Wat. Sci. Tech. 30(12): 55-61.
Broughton, M. J., Thiele,J.H. Birch,E.J. and Cohen,A. (1998). “Anaerobic batch digestion of sheep
tallow.” Wat. Res. 32(5): 1423-1428.
Bryant, M. P., Wollin, E.A., Wollin, M.J. and Wolfe, R.S. (1967). “Methanobacillus smellaszii, a
symbiotic association of two species of bacteria.” Arc. Microbiol. 59: 20-24.
187
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Charles, W., Ho, G. and Cord-Ruwisch, R. (1996). “ Anaerobic bioflocculation of wool scouring
effluent: the influent of non-ionic surfactant on efficiency.” Wat. Sci. Tech. 34: 1-8.
Chui, H.K. and. Fang, H.H.P. (1994). “Histological analysis of microstructure of UASB granules.”
J. Environ. Eng. 120(5): 1322-1325.
Conway de Macario, E. and Macario A.J. (1982). “Specification antisera and immunological
procedures for characterization of methanogen bacteria.” J. Bacteriol. 149: 320-328.
Demeyer, D.I. and Henderrickx, H.K. (1967). “The effect of C 18 unsaturated fatty acids on
methane production in vitro by mixed rumen bacteria.” Biochemistry ET BIOPHYSICA ACTA
137: 484-497.
Devereux, R., Kane, M.D., Winfrey, J. and Stahl, D.A. (1992). “Genus- and group-specific
hybridization probes for determinative and environmental studies of sulfate-reducing bacteria.”
Syst. Appl. Microbiol. 15(601-609).
Edelmann, W., Engeli, H. and Gradenecker, M. (2000). “Co-digestion of organic solid waste and
sludge from sewage treatment.” Wat. Sci. Tech. 41(3): 213-221.
El-mamouni, R., Leduc, R. and Guiot, S.R. (1997). “Influence of the starting microbial nucleus
type on the anaerobic granulation dynamics.” Appl. Microbiol. Biotechnol. 47: 189-194.
188
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
El-mamouni, R., Leduc,R. and Guiot,S.R. (1998). “Influence of synthetic and natural polymers on
the anaerobic granulation process.” Wat. Sci.Tech. 38(8-9): 341-347.
Fang, H. H. P., Chui, H.K. and Li, Y.Y. . (1994a). “Microbial structure and activity of UASB
granules treating different wastewaters.” Wat. Sci. Tech. 30: 87-96.
Fang, H. H. P., Chui, H.K. and Chen, T. (1994b). “Performance and granule characteristics of
UASB process treating wastewater with hydrolysed proteins.” Wat. Sci. Tech. 30(8): 55-63.
Farrell, J.B., Earlap, A.E., Rick Baugh, J. Freedman, D. and Hayes, S. (1988). “Influence of
feeding procedure on microbial reductions and performance of anaerobic digestion.” J. water
Pollut. Control Fed. 60: 635-642.
Fongastitkul, P. D. S. and Mavinic, K.V.L. (1994). “A two-phased anaerobic digestion process:
concept, process failure and maximum system loading rate.” Wat. Environ. Res. 66(3): 243-254.
Galbraith, H., Miller, T.B., Paton, A.M. and Thomson, J.K. (1971). “Antibacterial activity of long
chain fatty acids and reversal with calcium, magnesium, ergocalciferol and cholesterol.” J. Appl.
Bacteriol. 34: 803-813.
Galbraith, H. and Miller, T.B. (1973). “Physicochemical effects of long chain fatty acids on
bacterial cell and their protoplasts.” J. Appl. Bacteriol. 36: 647-658.
189
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Ghosh, S. and Pohland, F.G. (1974). " Kinetic of substrate assimilation and product formation in
anaerobic digestion. J. Water Poll.Control Fed., 46: 748-759.
Ghosh, S., Conrad, J.R. and Klass, D.L. (1975). “Anaerobic acidogenesis of wastewater sludge.” J.
Water Poll. Control Fed. 47; 30-45.
Grotenhuis, J. T. C. Smit, M. Plugge, C.M. Xu, y. Lammeren, A. Stam, A. and Zehder, A. (1991).
“Bacteriological composition and structure of granular sludge adapted to different substrates.”
Appl. Environ. Microbiol. July: 1942-1949.
Gutierrez, S., Hernandez, A. and Vinas, M. (1999). “Mechanism of degradation of wool wax in the
anaerobic treatment of wool scouring wastewater.” Wat. Sci. Tech. 40(8): 17-23.
Haandel, A.C. and. Lettinga, G. (1994). "Anaerobic sewage treatment", John Weley &Sons.
Hanaki, K., Ishikawa, T. and Matssumoto, J. (1983). “Inhibitory and stimulative effects of oleate
on methanogenesis from acetate in anaerobic digestion.” Technology Reports, Tohoku Univ.
48(No.2): 123-135.
Hanaki, K. (1981). “Mechanism of inhibition caused by long-chain fatty acids in anaerobic
digestion process.” Biotech. Bioeng. 23: 1591-1610.
190
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Hanaki, K., Matsuo, T. Nagase, M. and Tabata, Y. (1987). “Evaluation of effectiveness of twophase anaerobic digestion process degrading complex substrate.” Wat. Sci. Tech. 19: 311-322.
Hannsson, M., Nordberg, A., Sundh, I. and Mathisen, B. (2001). “Early warning of disturbances in
a laboratory scale MSW biogas process.” Anaerobic Digestion 2001 part1:,Belgium, 295-300.
Hansen, K.H., Ahring, B.K. and Raskin, L. (1999). “Quantification of syntrophic fatty acid-Boxidizing bacteria in a mesophilic biogas reactor by oligonucleotide probe hybridization.” Appl.
Environ. Microbiol. Nov.: 4767-4774.
Harmsen, H.J.M., Kengen, H.M.P., Akkermans, A.D.L., Stams, A.J.M. and de Vos,W.M. (1996).
“Detectation and localization of syntrophic propionate-oxidizing bacteria in granular sludge by InSitu hybridization using 16SrRNA-based oligonucleotide probes.” Appl. Environ. Microbiol.
62(5): 1656-1663.
Harmsen, H.J.M., Akkermans, A.D.L., Stams, A.J.M. and de Vos,W.M. (1996). “Population
dynamics of propionate-oxidizing bacteria under methanogenic and sulfidogenic conditions in
anaerobic granular sludge.” Appl. Environ. Microbiol. 62(6): 2163-2168.
Hicks, R.E., Amann, R.I. and Stahl, D.A. (1992). “Dual staining of natural bacterioplankton with
4, 6-diamidino-2-phenylinodole and fluorescent oligonucleotide probes targeting kingdom-level
16S rRNA sequences.” Appl. Environ. Microbiol. 58(7): 2158-2163.
191
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Hwu, C.-S., Donlon, B. and Lettinga, G. (1996). “Comparative toxicity of long-chain fatty acid to
anaerobic sludge from various origins.” Wat. Sci. Tech. 34: 5-6.
Hwu, C.S. (1997). "Enhancing anaerobic treatment of wastewaters containing oleic acid",
Wageningen Agricultural University, Wageningen , The Netherlands.
Hwu, C.S., Tseng, S.K., Yuan, C.Y., Kulik, Z. and Lettinga, G. (1998). “Biosorption of long-chain
fatty acids in UASB treatment process.” Wat. Res. 32: 1571-1579.
Ibrahim, A., Yeoh, B.G., Cheah, S.C., Ma, A., Ahmad, S., Chew, T.Y., Rag, R. and Wahid, M.J.A.
(1984). “Thermophilic anaerobic contact digestion of palm oil mill effluent.” Wat. Sci.Tech. 17:
155-166.
Imai, T., Ukita, M. Liu, J. Sekine, M., Nakannishi, H. and Fukagawa, M. (1997). “Advanced start
up of UASB reactors by adding of water absorbing polymer.” Wat. Sci.Tech. 36(6-7): 399-406.
Jeris, J. S. and McCarty, M. (1965). “The biochemistry of methane fermentation using 14 C
tracers.” J. Water Pollut.Control Fed. 37: 178-192.
Jeyaseelan, S. and McCarty, T. (1995). “Effects of phase separation in anaerobic on different
substrates.” Wat. Sci.Tech. 31(9): 153-162.
192
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Kane, M.D., Poulsen, L.K. and Stahl, D.A. (1993). “Monitoring the enrichment and isolation of
sulfate-reducing bacteria by using oligonucleotide hybridization probes designed from
environmentally derived 16S rRNA sequences.” Appl. Environ. Microbiol. 59(3): 682-686.
Kleerebezem, R., Mortier, J., Hulshoff, L.W. and Lettinga, G. (1997). “Anaerobic Pre-treatment of
petrochemical effluents: terephthalic acid.” Wat. Sci.Tech. 36(2-3): 237-248.
Kodicek, E. and. Worden, A.N. (1945). “The effect of unsaturated fatty acids on L. Helcveticus
and other gram-positive micro-organisms.” Biochem. J. 39: 78-85.
Komastsu, T., Hanaki, K. and Matsuo,T. (1991). "Prevention of lipid inhibition in anaerobic
processes by introducing a two-phase system." Wat. Sci. Tech. 23 1189-1200
Koster, I. W., and Cramer, A. (1986). “Inhibition of methanogenesis from acetate in granular
sludge by long-chain fatty acid.” Appl. Environ. Microbiol. 53: 403-409.
Koster, I. W. (1987). “Abatement of long-chain acid inhibition of methanogenesis by calcium
addition.” Biological Wastes 22 (1987): 295-301.
Kubler, H., Hoppenheidt, K., Hirsch,P., Kottmair, A., Nimmrichter, R., Mucke, W. and Swerev,
M. (2000). “Full scale co-digestion of organic waste.” Wat. Sci. Tech. 41(3): 195-202.
Kuang,
193
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Lapsirikul, W., Cord-Ruwisch ,R.and Ho, G. (1993). “Anaerobic bioflocculation of wool scouring
effluent.” Wat. Res. 28: 1743-1747.
Lettinga, G., Velsan, A.F.M., Hobma, S.W., Zeeuw, W. and Klapwijk, A. (1980). “Use of UASB
reactor concept for biological wastewater treatment.” Biotech. Bioeng. 22(4): 699.
Lettinga, G., Field, J., Lier, J., Zeeman, G. and Hulshoff-Pol, L.W. (1997). “Advanced anaerobic
wastewater treatment in the near future.” Wat. Sci. Tech. 35(10): 5-12.
Lier, J. B., Rebac, S., Lens, P., Bijnen, F, Elferink, S., Stams, A. and Lettinga, G. (1997).
“Anaerobic treatment of partly acidified wastewater in a two-stage expanded granular sludge bed
(EGSB) system at 8 o C.” Wat. Sci. Tech. 36(6-7): 317-324.
Lier, J. B., Ahring, A.B.K., Macarie, H., Moletta, R., Dohanyos, M., Hushoff Pol, R. L.W., Lens,
P. and Vestraete,W. (2001). “New perspectives in anaerobic digestion.” Wat. Sci. Tech 43(1): 1-
Londry, K. L. (1997). “Anaerobic degradation of m-cresol by a sulfate-reducing bacterium.” Appl.
Environ. Microbiol 63(8): 3170-3175.
Lorowitz, W. H., Zhao, H. and Bryant, M.P. (1989). “Syntrophomonas wolfei sub. sp. saponavida
sub sp. nov., a long chain fatty acids degradating, anaerobic, syntrophic bacterium; and emended
descriptions of the genus and species.” Intl. J. Syst. Bacteriol. 39: 122-126.
194
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Lovley, D. R., Dwyer, D. F. and Klug, M. J. (1982). "Kinetic Analysis of competition Between
Sulfate Reducers and Methanogens for Hydrogens in Sediments." Appl. Environ. Microbiol. Mar:
1982
Macleod, F. A., Guiot, S.R. and Costerton, J.W. (1990). “Layered structure of bacterial aggregates
produced in an upflow anaerobic sludge bed and filter reactor.” Appl. Environ. Microbiol. June:
1598-1607.
Mah, R. T. and Smith, M. R. (1986). "The Prokaryotes. The methanogenic bacteria." R. A. Mah
and M. R. Smith, New Nork.
Manz, W., Wagner, M.,
Amann, R.,
Schleifer, K. (1993). “In situ characterization of the
microbial consortia active in two wastewater treatment plants.” Wat. Res. 43: 1715-1723.
McAvoy, D. C., Shimp, R.J., Namkung, E. and Hand, V.C. (1996). “Fate and effects of olestra, a
fat substitute, during conventional wastewater treatment.” Wat. Environ. Res. 68: 169-177.
McCarty, P. L. and Mosey, F.E. (1991). “Modelling of anaerobic digestion processes.” Wat. Sci.
Tech. 24(8): 17-33.
McInerney, M. J., Mackie, R.I. and Bryant, M.P. (1981). “Synthrophic association of a butyratedegrading bacterium and Methanosarcina enriched from rumen fluid.” Appl. Environ. Microbiol.
41: 826-828.
195
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Miron, Y., Zeeman, G., Lier, J.B. and Lettinga, G. (2000). “The role of sludge retention time in the
hydrolysis and acidification of lipids, carbohydrates and proteins during digestion of primary
sludge in CSTR system.” Wat. Sci. Tech. 34(5): 1705-1713.
Mousa L. and Forster, C. F. (1999). “The use of glucose as a growth factor to counteract inhibition
in anaerobic digestion.” Trans IChemE 77 (partB): 193-198.
Mousa, L. and. Forster, C.F. (1998). “The effect of trace organics on the inhibition of gas
production by anaerobic sludges: batch studies.” Wat. Res. 32(12): 3795-3798.
Nieman, C. (1954). “Influence of trace amounts of fatty acids on the growth of microorganisms.”
Bacteriological reviews 18: 147-163.
Norman Pace (1997) "Biological Perspective on the Early Earth and the Chemistry that Spawned
Life"
http://astrobiology.arc.nasa.gov/workshops/1996/astrobiology/speakers/pace/pace_abstract.html
Novaes, R. F. V. (1986). “Microbiology of anaerobic digestion.” Wat. Sci. Tech. 18(12): 1-14.
Nunn, W. D. (1986). “A molecular view of fatty acid catabolism in Escherichia coli.”
Microbiological reviews 50: 179-192.
O'Flaherty, V., Lens, P.N.L., Beer, D. and Colleran, E. (1997). “Effect of feed composition and
upflow velocity on aggregate characteristics in anaerobic upflow reactors.” Appl
Biotechnol 47: 102-107.
196
Microbiol
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Pavlostathis, S. G. and Giraldogomez, E. (1991). “Kinetics of anaerobic treatment.” Wat. Sci. Tech
24(8): 35-59.
Petruy, R. and Lettinga, G. (1997). “Digestion of a milk-fat emulsion.” Bioresource Technol. 61:
141-149.
Palaeos:The trace of life on earth (2003).
(http://www.palaeos.com/Kingdoms/Prokaryotes/Archaea.htm)
Quemeneur, M. and Marty, Y. (1994). “Fatty acids and sterols in domestic wastewater.” Wat. Res.
28: 1217-1226.
Raskin, L., J.M. Rittmann, B.E. and Stahl, D.A. (1994a). “Group-specific 16S rRNA hybridization
probes to describe natural communities of methanogens.” Appl. Environ. Microbiol. 60(4): 12321240.
Raskin, L., Poulsen, L.K.,
Noguera , D.L.,
Rittmann, B.E. and Stahl, D.A. (1994b).
“Quantification of methanogenic groups in anaerobic biological reactors by oligonucleotide probe
hybridization.” Appl. Environ. Microbiol. 60(4): 1241-1248.
Raskin, L., Amann, R.I., Poulsen, L.K., Rittmann, B.E. and Stahl, D.A. (1995). “Use of ribosomal
RNA-based molecular probes for characterization of complex microbial communities in anaerobic
biofilms.” Wat. Sci. Tech. 31(1): 261-272.
197
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Raskin, L. Rittmann, B.E. and Stahl, D.A. (1996). “Competition and coexistence of Sulfatereducing and methanogenic populations in anaerobic biofilms.” Appl. Environ. Microbiol,
62(Oct.): 3847-3857.
Ratledge, C. (1994). "Biodegradation of oils, fats and fatty acids-Biochemistry of Microbial
degradation." Kluwer academic publ.
Raunkjjaer, K.,
Jacobsen, T. and Nielsen, K. (1994). " Measurement of pools of protein,
carbohydrate and lipid in domestic wastewater" Wat. Sci. Tech. 28 (2): 251-262
Rinzema, A. (1988). "Anaerobic treatment of wastewater with high concentrations of lipids or
sulfate". PhD thesis, Wageningen Agricultural University. Wageningen, The Netherlands.
Rinzema, A. (1993). “Anaerobic digestion of long-chain fatty acids in UASB and expanded
granular sludge bed reactors.” Process Biochemistry 28: 527-537.
Rinzema, A., Boone, M., Knippenberg, K. and Lettinga, G. (1994). “Bacterial effect of long chain
fatty acids in anaerobic digestion.” Wat. Environ. Res. 66: 40-49.
Rouviere, P.
Winker, S. and Woese, C.R. (1992). “A detailed phylogeny for the
methanomicrobiales.” System. Appl. Microbiol. 15: 363-371.
198
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Roy, F., Samain, E., Dubourguier, H.C. and Albagnac, G. (1986). “Syntrophoman as sapovorans
sp. nov., a new oblgately proton reducing anaerobe oxidizing saturated and unsaturated long chain
fatty acids.” Arch. Microbiol. 145: 142-147.
Rozzi, A. and Remigi, E. (2002). “Kinetics of the anammox bacteria by a pH-stat bioassay.”
Anaerobic Digestion 2001, Belgium. part1: 301-303.
Salminen, E., Rintala, J., Lokshina,Y.L., and Vavilin,V.A. (2000). “Anaerobic batch degradation
of solid poultry slaughterhouse waste.” Wat. Sci. Tech. 41(3): 33-41.
Sam-soon, P. A. L. N. S., Loewenthal, R.E., Dold, P.L. and Mariais, G.V.R. (1987). “Hypothesis
for pelletization in the upflow anaerobic sludge bed reactor.” Water SA. 13: 68-80.
Sam-soon, P. A. L. N. S., Loewenthal, R.E., Dold, P.L. and Mariais, G.V.R. (1991). “A long chain
fatty acid, oleate, as sole substrate in upflow anaerobic sludge bed (UASB) systems.” Water SA.
16: 165-170.
Sanders, W. T. M., Geerink, M., Zeeman, G. and Lettinga, G. (2000). “Anaerobic hydrolysis
kinetics of particulate substrates.” Wat. Sci. Tech. 41(3): 17-24.
Santegoeds, C. M., Damgaard, L.R., Hesselink, G., Zopei, J., Lens, P., Muyzer, G. and Beer, D.
(1999). “Distribution of sulfate reducing and methanogenic bacteria in anaerobic aggregates
determined by microsensor and molecular analyses.” Appl. Environ. Microbiol. Oct.: 4618-4629.
199
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Sayed, S. K. I. (1987). "Anaerobic treatment of slaughterhouse wastewater using the UASB
process". PhD thesis, The Agricultural University of Wageningen, Wageningen, The Netherlands
Schieder, D., Schneider, R. and Bischof, F. (2000). “Thermal hydrolysis (TDH) as a pretreatment
method for the digestion of organic waste.” Wat. Sci. Tech 41(3): 181-187.
Schmidt, J. E. and Ahring, K. (1996). “Granular sludge formation in upflow anaerobic sludge
blanket (UASB) reactors.” Biotechnol Bioeng 49: 229-246.
Sekiguchi, Y., Kamagata,Y., Nakamura, K., Ohashi, A. and Harada, H. (1999). “Fluorescence on
situ hybridization using 16S rRNA -Targeted oligonucleotides reveals location of methanogens
and selected uncultured bacteria in mesophilic and thermophilic sludge granules.” Appl. Environ.
Microbiol. Mar.: 1280-1288.
Sorensen, A. H. T., V.L. Torsvik, T. Poulsen, L.K. and Ahring, B.K. (1997). “Whole-cell
hybridization of Methanosarcina cells with two new oligonucleotide probes.” Appl. Environ.
Microbiol. 63(8): 3043-3050.
Stahl, D. A., Flesher, B., Mansfield, H.R. and Montgomery, L. (1988). “Use of phylogenetically
based hybridization probes for studies of ruminal microbial ecology.” Appl. Environ. Microbiol.
may: 1079-1084.
200
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Stahl, D. A. and Amann, R.I. (1991). "Development and application of nucleic acid probes". John
Willey and sons, New Nork.
Stoll, U. and Gupta, H. (1997). “Management strategies for oil and grease residues.” Waste
management & Research 15: 23-32.
Svetlitshnyi, V., Rainey, F., and Wiegel, J. (1996). “Thermosyntropha lipolytica gen. nov., sp.nov.,
a lipolytic, anaerobic alkalitolerant, thermophilic bacterium utilizing short- and long chain fatty
acids in sytrophic co-culture with a methanogenic archaeum.” Intl. J. Syst. Bacteriol. 46: 11311137.
Thies, E., Jenkins, T. and Stutzenberger, F. (1994). “Effects of detergent tween 80 on
Thermomonospora curvata.” World J. Syst. Bacteriol. 46: 1131-1137.
Toerien, D. F. and Hattingh, W.H.J. (1969). “The microbiology of anaerobic digestion.” Wat. Res.
3: 385-414.
Trevizo, C. and Nirmalakhandan, N. (1999). “Prediction of microbial toxicity of industrial organic
chemicals.” Wat. Sci. Tech. 39: 10-11.
Van Dyke, M.I and McCarthy, A.J. (2002). " Molecular biological detection and characterisation
of Clostridium population in municiple landfill sites." Appl. Environ. Microbiol., 68, 2049-2053.
201
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Viviani, R. (1970). “Metabolism of long-chain fatty acids in the rumen.” Advanced in lipid
research: 246-346.
Weng, C.N. and Jeris, J.S. (1976). “Biochemical mechanisms in the methane fermentation of
glutamic and oleic acids.” Wat. Res. 10: 9-18.
Wentzel, M.C., Moosbrugger, R.E., Sam-soon, P.A.L.N., Ekama, G.A. and Marais, G.R. (1994).
“Tentative guidelines for waste selection, process design, operation and control of upflow
anaerobic sludge bed reactors.” Wat. Sci. Tech. 30(12): 31-42.
Woese, C.R., Kandler, O. and Wheelis, M.L. (1990). “Towards a natural system of organisms:
proposal for domains Archaea, Bacteria, and Eucarya.” Proc. Natl. Acad. Sci. USA 87: 4576-4579.
Wu, W. (1996). “Formation of fatty-acid-degrading, anaerobic granules by defined species.” Appl.
Environ. Microbiol. 62(6): 2037-2044.
Yan, G. and Allen, D.G. (1994). “Biosorption of high molecular weight organochlorines in pulp
mill effluent.” Wat. Res. 28: 1933-1941.
Yilmazer, G. and. Yenigun, O. (1999). “Two-phase Anaerobic treatment of cheese whey.” Wat.
Sci. Tech. 40(1): 289-295.
202
Enhancing Anaerobic Degradation of Lipids in Wastewater by Addition of Substrate - Y. Kuang
REFERENCE
Zeeman, G. and Sanders, W. (2001)."Potential of anaerobic digestion of complex waste (water)."
Wat. Sci. Tech. 44 (8):115-122
Zeikus, J. G. (1980). "Microbial populations in digesters". Applied Sci. Publishers, London.
Zhang, T. C. and Noike, T. (1991). “Comparison of one-phase and two-phase anaerobic digestion
processes in characteristics of substrate degradation and bacterial population levels.” Wat.
Sci.Tech. 23: 1157-1166.
Zhao, H., Yang, D. Woese, C.R. and Bryant, M.P. (1993). “Assignment of fatty acid-b-oxidizing
sytrophic bacteria to Sytrophomonadaceae fam.nov. on the basis of 16S rRNA sequence analyses.”
J. Syst. Microbiol. 40: 40-44.
Zheng, D., Alm, E.W., Stahl, D.A. and Raskin, L. (1996). “Characterization of universal smallsubunit rRNA hybridization probes for quantitative molecular microbial ecology studies.” Appl.
Environ. Microbiol. 62(12): 4504-4513.
Zhu, J. Hu, J. and Gu, X. (1997). “The bacterial numeration and an observation of a new
syntrophic association for granular sludge.” Wat. Sci. Tech. 36(6-7): 133-140.
Zinder, S. H., Anguise, T. and Cardwell, S.C. (1984). “Effects of temperature on methanogenesis
in thermophilic (580) anaerobic digester C.” Appl. Environ. Microbiol. 47: 808-813.
203